doc_key stringlengths 10 20 | ner listlengths 0 22 | ner_mapping_to_source listlengths 0 22 | relations listlengths 0 22 | relations_mapping_to_source listlengths 0 22 | sentence listlengths 1 93 | sentence-detokenized stringlengths 1 494 | token2charspan listlengths 1 93 |
|---|---|---|---|---|---|---|---|
ai-train-1 | [
[
4,
10,
"product"
],
[
17,
19,
"field"
],
[
22,
24,
"task"
],
[
27,
29,
"task"
],
[
34,
37,
"task"
],
[
41,
42,
"field"
],
[
43,
44,
"researcher"
],
[
46,
47,
"researcher"
],
[
49,
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13
] | [
[
4,
10,
17,
19,
"part-of",
"",
false,
false
],
[
4,
10,
17,
19,
"usage",
"",
false,
false
],
[
4,
10,
22,
24,
"part-of",
"",
false,
false
],
[
4,
10,
22,
24,
"usage",
"",
false,
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9
] | [
"Les",
"approches",
"populaires",
"du",
"système",
"de",
"recommandation",
"basé",
"sur",
"l'",
"opinion",
"utilisent",
"diverses",
"techniques",
",",
"notamment",
"l'",
"exploration",
"de",
"texte",
",",
"la",
"recherche",
"d'",
"informations",
",",
"l'",
"anal... | Les approches populaires du système de recommandation basé sur l'opinion utilisent diverses techniques, notamment l'exploration de texte, la recherche d'informations, l'analyse des sentiments (voir également l'analyse multimodale des sentiments) et l'apprentissage profond X.Y. Feng, H. Zhang, Y.J. Ren, P.H. Shang, Y. Z... | [
[
0,
3
],
[
4,
13
],
[
14,
24
],
[
25,
27
],
[
28,
35
],
[
36,
38
],
[
39,
53
],
[
54,
58
],
[
59,
62
],
[
63,
65
],
[
65,
72
],
[
73,
82
],
[
83,
91
],
[
9... |
ai-train-2 | [
[
9,
9,
"university"
],
[
15,
16,
"researcher"
],
[
18,
19,
"researcher"
]
] | [
0,
1,
2
] | [
[
15,
16,
9,
9,
"physical",
"",
false,
false
],
[
15,
16,
9,
9,
"role",
"",
false,
false
],
[
18,
19,
9,
9,
"physical",
"",
false,
false
],
[
18,
19,
9,
9,
"role",
"",
false,
f... | [
0,
1,
2,
3
] | [
"Les",
"défenseurs",
"des",
"représentations",
"procédurales",
"étaient",
"principalement",
"centrés",
"au",
"MIT",
",",
"sous",
"la",
"direction",
"de",
"Marvin",
"Minsky",
"et",
"Seymour",
"Papert",
"."
] | Les défenseurs des représentations procédurales étaient principalement centrés au MIT, sous la direction de Marvin Minsky et Seymour Papert. | [
[
0,
3
],
[
4,
14
],
[
15,
18
],
[
19,
34
],
[
35,
47
],
[
48,
55
],
[
56,
70
],
[
71,
78
],
[
79,
81
],
[
82,
85
],
[
85,
86
],
[
87,
91
],
[
92,
94
],
[
9... |
ai-train-3 | [
[
12,
12,
"programlang"
]
] | [
0
] | [] | [] | [
"L'",
"interface",
"standard",
"et",
"l'",
"interface",
"de",
"la",
"calculatrice",
"sont",
"écrites",
"en",
"Java",
"."
] | L'interface standard et l'interface de la calculatrice sont écrites en Java. | [
[
0,
2
],
[
2,
11
],
[
12,
20
],
[
21,
23
],
[
24,
26
],
[
26,
35
],
[
36,
38
],
[
39,
41
],
[
42,
54
],
[
55,
59
],
[
60,
67
],
[
68,
70
],
[
71,
75
],
[
7... |
ai-train-4 | [
[
0,
0,
"product"
],
[
32,
32,
"product"
]
] | [
0,
1
] | [
[
0,
0,
32,
32,
"related-to",
"compatible_with",
false,
false
]
] | [
0
] | [
"Octave",
"permet",
"de",
"résoudre",
"numériquement",
"des",
"problèmes",
"linéaires",
"et",
"non",
"linéaires",
",",
"et",
"d'",
"effectuer",
"d'",
"autres",
"expériences",
"numériques",
"à",
"l'",
"aide",
"d'",
"un",
"logiciel",
"qui",
"est",
"en",
"grande"... | Octave permet de résoudre numériquement des problèmes linéaires et non linéaires, et d'effectuer d'autres expériences numériques à l'aide d'un logiciel qui est en grande partie compatible avec MATLAB. | [
[
0,
6
],
[
7,
13
],
[
14,
16
],
[
17,
25
],
[
26,
39
],
[
40,
43
],
[
44,
53
],
[
54,
63
],
[
64,
66
],
[
67,
70
],
[
71,
80
],
[
80,
81
],
[
82,
84
],
[
8... |
ai-train-5 | [
[
4,
6,
"algorithm"
],
[
10,
12,
"misc"
],
[
14,
15,
"researcher"
],
[
22,
24,
"university"
]
] | [
0,
1,
2,
3
] | [
[
4,
6,
14,
15,
"origin",
"",
false,
false
],
[
10,
12,
14,
15,
"origin",
"",
false,
false
],
[
14,
15,
22,
24,
"physical",
"",
false,
false
],
[
14,
15,
22,
24,
"role",
"",
false,... | [
0,
1,
2,
3
] | [
"Des",
"variantes",
"de",
"l'",
"algorithme",
"de",
"rétropropagation",
"ainsi",
"que",
"les",
"méthodes",
"non",
"supervisées",
"de",
"Geoff",
"Hinton",
"et",
"de",
"ses",
"collègues",
"de",
"l'",
"université",
"de",
"Toronto",
"peuvent",
"être",
"utilisées",
... | Des variantes de l'algorithme de rétropropagation ainsi que les méthodes non supervisées de Geoff Hinton et de ses collègues de l'université de Toronto peuvent être utilisées pour former des architectures neuronales profondes et hautement non linéaires, {{cite journal | [
[
0,
3
],
[
4,
13
],
[
14,
16
],
[
17,
19
],
[
19,
29
],
[
30,
32
],
[
33,
49
],
[
50,
55
],
[
56,
59
],
[
60,
63
],
[
64,
72
],
[
73,
76
],
[
77,
88
],
[
8... |
ai-train-6 | [
[
8,
8,
"metrics"
]
] | [
0
] | [] | [] | [
"ou",
"de",
"manière",
"équivalente",
"en",
"utilisant",
"la",
"notation",
"DCG",
":"
] | ou de manière équivalente en utilisant la notation DCG : | [
[
0,
2
],
[
3,
5
],
[
6,
13
],
[
14,
25
],
[
26,
28
],
[
29,
38
],
[
39,
41
],
[
42,
50
],
[
51,
54
],
[
55,
56
]
] |
ai-train-7 | [
[
0,
2,
"algorithm"
],
[
6,
8,
"algorithm"
],
[
16,
19,
"algorithm"
],
[
23,
27,
"algorithm"
],
[
31,
31,
"algorithm"
],
[
33,
35,
"algorithm"
],
[
52,
53,
"misc"
]
] | [
0,
1,
2,
3,
4,
5,
6
] | [
[
0,
2,
6,
8,
"type-of",
"",
false,
false
],
[
0,
2,
16,
19,
"usage",
"part-of?",
true,
false
],
[
16,
19,
23,
27,
"compare",
"",
false,
false
],
[
31,
31,
23,
27,
"part-of",
"",
f... | [
0,
1,
2,
3
] | [
"Les",
"cartes",
"auto-organisatrices",
"diffèrent",
"des",
"autres",
"réseaux",
"neuronaux",
"artificiels",
"dans",
"la",
"mesure",
"où",
"elles",
"appliquent",
"un",
"apprentissage",
"compétitif",
",",
"par",
"opposition",
"à",
"l'",
"apprentissage",
"par",
"corre... | Les cartes auto-organisatrices diffèrent des autres réseaux neuronaux artificiels dans la mesure où elles appliquent un apprentissage compétitif, par opposition à l'apprentissage par correction d'erreurs (comme la rétropropagation avec descente de gradient), et dans la mesure où elles utilisent une fonction de voisinag... | [
[
0,
3
],
[
4,
10
],
[
11,
30
],
[
31,
40
],
[
41,
44
],
[
45,
51
],
[
52,
59
],
[
60,
69
],
[
70,
81
],
[
82,
86
],
[
87,
89
],
[
90,
96
],
[
97,
99
],
[
1... |
ai-train-8 | [
[
12,
14,
"organisation"
],
[
30,
31,
"misc"
],
[
39,
43,
"metrics"
]
] | [
0,
1,
2
] | [] | [] | [
"Depuis",
"le",
"début",
"des",
"années",
"1990",
",",
"plusieurs",
"organismes",
",",
"dont",
"l'",
"Audio",
"Engineering",
"Society",
",",
"recommandent",
"que",
"les",
"mesures",
"de",
"la",
"gamme",
"dynamique",
"soient",
"effectuées",
"en",
"présence",
"d... | Depuis le début des années 1990, plusieurs organismes, dont l'Audio Engineering Society, recommandent que les mesures de la gamme dynamique soient effectuées en présence d'un signal audio, qui est ensuite filtré dans la mesure du plancher de bruit utilisée pour déterminer la gamme dynamique. Cela permet d'éviter les me... | [
[
0,
6
],
[
7,
9
],
[
10,
15
],
[
16,
19
],
[
20,
26
],
[
27,
31
],
[
31,
32
],
[
33,
42
],
[
43,
53
],
[
53,
54
],
[
55,
59
],
[
60,
62
],
[
62,
67
],
[
68... |
ai-train-9 | [
[
6,
7,
"misc"
],
[
21,
23,
"task"
],
[
25,
29,
"task"
],
[
31,
32,
"task"
],
[
34,
35,
"task"
],
[
39,
40,
"task"
],
[
37,
46,
"task"
],
[
48,
52,
"field"
]
] | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
[
6,
7,
21,
23,
"part-of",
"concept_used_in",
true,
false
],
[
6,
7,
25,
29,
"part-of",
"concept_used_in",
false,
false
],
[
6,
7,
31,
32,
"part-of",
"concept_used_in",
false,
false
],
[
6,
7,
34,... | [
0,
1,
2,
3,
4,
5,
6
] | [
"La",
"technique",
"utilisée",
"pour",
"créer",
"des",
"visages",
"propres",
"et",
"les",
"utiliser",
"pour",
"la",
"reconnaissance",
"est",
"également",
"utilisée",
"en",
"dehors",
"de",
"la",
"reconnaissance",
"des",
"visages",
":",
"reconnaissance",
"de",
"l'... | La technique utilisée pour créer des visages propres et les utiliser pour la reconnaissance est également utilisée en dehors de la reconnaissance des visages : reconnaissance de l'écriture manuscrite, lecture labiale, reconnaissance vocale, interprétation du langage des signes/des gestes des mains et analyse de l'image... | [
[
0,
2
],
[
3,
12
],
[
13,
21
],
[
22,
26
],
[
27,
32
],
[
33,
36
],
[
37,
44
],
[
45,
52
],
[
53,
55
],
[
56,
59
],
[
60,
68
],
[
69,
73
],
[
74,
76
],
[
7... |
ai-train-10 | [
[
1,
3,
"organisation"
],
[
6,
10,
"organisation"
],
[
12,
12,
"organisation"
],
[
16,
20,
"organisation"
],
[
23,
27,
"organisation"
],
[
30,
34,
"organisation"
],
[
37,
41,
"organisation"
],
[
43,
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
[
6,
10,
1,
3,
"part-of",
"",
false,
false
],
[
12,
12,
6,
10,
"named",
"",
false,
false
],
[
16,
20,
1,
3,
"part-of",
"",
false,
false
],
[
23,
27,
1,
3,
"part-of",
"",
false,
... | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
"La",
"National",
"Science",
"Foundation",
"chapeautait",
"la",
"National",
"Aeronautics",
"and",
"Space",
"Administration",
"(",
"NASA",
")",
",",
"le",
"ministère",
"américain",
"de",
"l'",
"énergie",
",",
"le",
"ministère",
"américain",
"du",
"commerce",
"NIS... | La National Science Foundation chapeautait la National Aeronautics and Space Administration (NASA), le ministère américain de l'énergie, le ministère américain du commerce NIST, le ministère américain de la défense, la Defense Advanced Research Projects Agency (DARPA) et l'Office of Naval Research, qui coordonnaient de... | [
[
0,
2
],
[
3,
11
],
[
12,
19
],
[
20,
30
],
[
31,
42
],
[
43,
45
],
[
46,
54
],
[
55,
66
],
[
67,
70
],
[
71,
76
],
[
77,
91
],
[
92,
93
],
[
93,
97
],
[
9... |
ai-train-11 | [
[
8,
12,
"metrics"
],
[
13,
14,
"algorithm"
],
[
19,
20,
"researcher"
],
[
26,
26,
"researcher"
]
] | [
0,
1,
2,
3
] | [
[
8,
12,
13,
14,
"part-of",
"",
false,
false
],
[
19,
20,
26,
26,
"related-to",
"added_appendix_to_work_of",
false,
false
]
] | [
0,
1
] | [
"Une",
"méthode",
"rapide",
"de",
"calcul",
"des",
"estimations",
"du",
"maximum",
"de",
"vraisemblance",
"pour",
"le",
"modèle",
"probit",
"a",
"été",
"proposée",
"par",
"Ronald",
"Fisher",
"en",
"annexe",
"du",
"travail",
"de",
"Bliss",
"en",
"1935",
"."
] | Une méthode rapide de calcul des estimations du maximum de vraisemblance pour le modèle probit a été proposée par Ronald Fisher en annexe du travail de Bliss en 1935. | [
[
0,
3
],
[
4,
11
],
[
12,
18
],
[
19,
21
],
[
22,
28
],
[
29,
32
],
[
33,
44
],
[
45,
47
],
[
48,
55
],
[
56,
58
],
[
59,
72
],
[
73,
77
],
[
78,
80
],
[
8... |
ai-train-12 | [
[
10,
11,
"product"
],
[
14,
15,
"product"
],
[
20,
20,
"organisation"
],
[
18,
18,
"product"
],
[
25,
25,
"organisation"
],
[
23,
23,
"product"
]
] | [
0,
1,
2,
3,
4,
5
] | [
[
18,
18,
14,
15,
"usage",
"uses_software",
false,
false
],
[
18,
18,
20,
20,
"artifact",
"",
false,
false
],
[
18,
18,
23,
23,
"named",
"",
false,
false
],
[
23,
23,
25,
25,
"artifact",
... | [
0,
1,
2,
3
] | [
"Plusieurs",
"de",
"ces",
"programmes",
"sont",
"disponibles",
"en",
"ligne",
",",
"comme",
"Google",
"Translate",
"et",
"le",
"système",
"SYSTRAN",
"qui",
"alimente",
"BabelFish",
"d'",
"AltaVista",
"(",
"devenu",
"Babelfish",
"de",
"Yahoo",
"depuis",
"le",
"... | Plusieurs de ces programmes sont disponibles en ligne, comme Google Translate et le système SYSTRAN qui alimente BabelFish d'AltaVista (devenu Babelfish de Yahoo depuis le 9 mai 2008). | [
[
0,
9
],
[
10,
12
],
[
13,
16
],
[
17,
27
],
[
28,
32
],
[
33,
44
],
[
45,
47
],
[
48,
53
],
[
53,
54
],
[
55,
60
],
[
61,
67
],
[
68,
77
],
[
78,
80
],
[
... |
ai-train-13 | [
[
3,
3,
"researcher"
],
[
6,
7,
"researcher"
],
[
9,
10,
"researcher"
],
[
21,
23,
"field"
],
[
27,
28,
"misc"
],
[
32,
34,
"field"
]
] | [
0,
1,
2,
3,
4,
5
] | [
[
3,
3,
21,
23,
"related-to",
"",
true,
false
],
[
3,
3,
27,
28,
"related-to",
"",
true,
false
],
[
3,
3,
32,
34,
"related-to",
"",
true,
false
],
[
6,
7,
21,
23,
"related-to",
"",
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
"En",
"2002",
",",
"Hutter",
",",
"avec",
"Jürgen",
"Schmidhuber",
"et",
"Shane",
"Legg",
",",
"a",
"développé",
"et",
"publié",
"une",
"théorie",
"mathématique",
"de",
"l'",
"intelligence",
"générale",
"artificielle",
"basée",
"sur",
"des",
"agents",
"intelli... | En 2002, Hutter, avec Jürgen Schmidhuber et Shane Legg, a développé et publié une théorie mathématique de l'intelligence générale artificielle basée sur des agents intelligents idéalisés et un apprentissage par renforcement motivé par la récompense. | [
[
0,
2
],
[
3,
7
],
[
7,
8
],
[
9,
15
],
[
15,
16
],
[
17,
21
],
[
22,
28
],
[
29,
40
],
[
41,
43
],
[
44,
49
],
[
50,
54
],
[
54,
55
],
[
56,
57
],
[
58,
... |
ai-train-14 | [
[
11,
11,
"metrics"
],
[
13,
17,
"metrics"
]
] | [
0,
1
] | [
[
11,
11,
13,
17,
"named",
"",
false,
false
]
] | [
0
] | [
"La",
"méthode",
"la",
"plus",
"courante",
"consiste",
"à",
"utiliser",
"la",
"mesure",
"dite",
"ROUGE",
"(",
"Recall-Oriented",
"Understudy",
"for",
"Gisting",
"Evaluation",
")",
"."
] | La méthode la plus courante consiste à utiliser la mesure dite ROUGE (Recall-Oriented Understudy for Gisting Evaluation). | [
[
0,
2
],
[
3,
10
],
[
11,
13
],
[
14,
18
],
[
19,
27
],
[
28,
36
],
[
37,
38
],
[
39,
47
],
[
48,
50
],
[
51,
57
],
[
58,
62
],
[
63,
68
],
[
69,
70
],
[
7... |
ai-train-15 | [
[
0,
0,
"product"
],
[
21,
21,
"programlang"
],
[
23,
23,
"programlang"
],
[
25,
26,
"researcher"
],
[
28,
29,
"organisation"
]
] | [
0,
1,
2,
3,
4
] | [
[
0,
0,
21,
21,
"related-to",
"",
false,
false
],
[
0,
0,
23,
23,
"related-to",
"",
false,
false
],
[
25,
26,
28,
29,
"role",
"",
false,
false
]
] | [
0,
1,
2
] | [
"RapidMiner",
"fournit",
"des",
"schémas",
"d'",
"apprentissage",
",",
"des",
"modèles",
"et",
"des",
"algorithmes",
"et",
"peut",
"être",
"étendu",
"à",
"l'",
"aide",
"de",
"scripts",
"R",
"et",
"Python",
".",
"David",
"Norris",
",",
"Bloor",
"Research",
... | RapidMiner fournit des schémas d'apprentissage, des modèles et des algorithmes et peut être étendu à l'aide de scripts R et Python. David Norris, Bloor Research, 13 novembre 2013. | [
[
0,
10
],
[
11,
18
],
[
19,
22
],
[
23,
30
],
[
31,
33
],
[
33,
46
],
[
46,
47
],
[
48,
51
],
[
52,
59
],
[
60,
62
],
[
63,
66
],
[
67,
78
],
[
79,
81
],
[
... |
ai-train-16 | [
[
1,
3,
"product"
],
[
9,
9,
"programlang"
],
[
11,
12,
"product"
]
] | [
0,
4,
5
] | [
[
1,
3,
11,
12,
"related-to",
"",
true,
false
],
[
11,
12,
9,
9,
"general-affiliation",
"",
true,
false
]
] | [
2,
4
] | [
"La",
"version",
"la",
"plus",
"récente",
",",
"entièrement",
"basée",
"sur",
"Java",
"(",
"Weka",
"3",
")",
",",
"dont",
"le",
"développement",
"a",
"débuté",
"en",
"1997",
",",
"est",
"aujourd'hui",
"utilisée",
"dans",
"de",
"nombreux",
"domaines",
"d'",... | La version la plus récente, entièrement basée sur Java (Weka 3), dont le développement a débuté en 1997, est aujourd'hui utilisée dans de nombreux domaines d'application différents, en particulier à des fins éducatives et de recherche. | [
[
0,
2
],
[
3,
10
],
[
11,
13
],
[
14,
18
],
[
19,
26
],
[
26,
27
],
[
28,
39
],
[
40,
45
],
[
46,
49
],
[
50,
54
],
[
55,
56
],
[
56,
60
],
[
61,
62
],
[
6... |
ai-train-17 | [
[
0,
1,
"product"
],
[
17,
24,
"misc"
],
[
28,
31,
"misc"
],
[
34,
42,
"conference"
]
] | [
0,
1,
2,
3
] | [
[
17,
24,
0,
1,
"topic",
"",
false,
false
],
[
17,
24,
28,
31,
"win-defeat",
"",
false,
false
],
[
28,
31,
34,
42,
"temporal",
"",
true,
false
]
] | [
0,
1,
2
] | [
"Eurisko",
"a",
"fait",
"de",
"nombreuses",
"découvertes",
"intéressantes",
"et",
"a",
"bénéficié",
"d'",
"une",
"grande",
"reconnaissance",
",",
"son",
"article",
"Heuretics",
":",
"Theoretical",
"and",
"Study",
"of",
"Heuristic",
"Rules",
"a",
"remporté",
"le"... | Eurisko a fait de nombreuses découvertes intéressantes et a bénéficié d'une grande reconnaissance, son article Heuretics : Theoretical and Study of Heuristic Rules a remporté le prix du meilleur article à l'Association for the Advancement of Artificial Intelligence de 1982. | [
[
0,
7
],
[
8,
9
],
[
10,
14
],
[
15,
17
],
[
18,
28
],
[
29,
40
],
[
41,
54
],
[
55,
57
],
[
58,
59
],
[
60,
69
],
[
70,
72
],
[
72,
75
],
[
76,
82
],
[
83... |
ai-train-18 | [
[
8,
10,
"metrics"
]
] | [
0
] | [] | [] | [
"Pour",
"tenir",
"compte",
"des",
"entités",
"multiples",
",",
"une",
"perte",
"de",
"charnière",
"distincte",
"est",
"calculée",
"pour",
"chaque",
"capsule",
"."
] | Pour tenir compte des entités multiples, une perte de charnière distincte est calculée pour chaque capsule. | [
[
0,
4
],
[
5,
10
],
[
11,
17
],
[
18,
21
],
[
22,
29
],
[
30,
39
],
[
39,
40
],
[
41,
44
],
[
45,
50
],
[
51,
53
],
[
54,
63
],
[
64,
73
],
[
74,
77
],
[
7... |
ai-train-19 | [
[
8,
10,
"product"
],
[
12,
13,
"product"
],
[
15,
16,
"product"
],
[
18,
19,
"product"
],
[
21,
23,
"product"
],
[
32,
33,
"product"
],
[
40,
45,
"product"
],
[
48,
49,
"product"
],
[
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
[
8,
10,
32,
33,
"type-of",
"",
false,
false
],
[
12,
13,
32,
33,
"type-of",
"",
false,
false
],
[
15,
16,
32,
33,
"type-of",
"",
false,
false
],
[
18,
19,
32,
33,
"type-of",
"",
f... | [
0,
1,
2,
3,
4,
5,
6
] | [
"Avec",
"l'",
"émergence",
"d'",
"assistants",
"conversationnels",
"tels",
"que",
"Siri",
"d'",
"Apple",
",",
"Amazon",
"Alexa",
",",
"Google",
"Assistant",
",",
"Microsoft",
"Cortana",
"et",
"Bixby",
"de",
"Samsung",
",",
"il",
"est",
"désormais",
"possible",... | Avec l'émergence d'assistants conversationnels tels que Siri d'Apple, Amazon Alexa, Google Assistant, Microsoft Cortana et Bixby de Samsung, il est désormais possible d'accéder aux portails vocaux via des appareils mobiles et des haut-parleurs intelligents de type Far Field tels que Amazon Echo et Google Home. | [
[
0,
4
],
[
5,
7
],
[
7,
16
],
[
17,
19
],
[
19,
29
],
[
30,
46
],
[
47,
51
],
[
52,
55
],
[
56,
60
],
[
61,
63
],
[
63,
68
],
[
68,
69
],
[
70,
76
],
[
77,... |
ai-train-20 | [
[
3,
4,
"field"
],
[
7,
9,
"algorithm"
],
[
12,
16,
"algorithm"
],
[
19,
21,
"algorithm"
],
[
24,
24,
"algorithm"
]
] | [
0,
1,
2,
3,
4
] | [
[
7,
9,
3,
4,
"type-of",
"",
false,
false
],
[
12,
16,
3,
4,
"type-of",
"",
false,
false
],
[
19,
21,
3,
4,
"type-of",
"",
false,
false
],
[
24,
24,
3,
4,
"type-of",
"",
false,
... | [
0,
1,
2,
3
] | [
"Des",
"exemples",
"d'",
"apprentissage",
"supervisé",
"sont",
"le",
"classificateur",
"Naive",
"Bayes",
",",
"la",
"machine",
"à",
"vecteurs",
"de",
"support",
",",
"les",
"mélanges",
"de",
"gaussiens",
"et",
"le",
"réseau",
"."
] | Des exemples d'apprentissage supervisé sont le classificateur Naive Bayes, la machine à vecteurs de support, les mélanges de gaussiens et le réseau. | [
[
0,
3
],
[
4,
12
],
[
13,
15
],
[
15,
28
],
[
29,
38
],
[
39,
43
],
[
44,
46
],
[
47,
61
],
[
62,
67
],
[
68,
73
],
[
73,
74
],
[
75,
77
],
[
78,
85
],
[
8... |
ai-train-21 | [
[
4,
5,
"algorithm"
],
[
28,
32,
"algorithm"
],
[
35,
35,
"task"
],
[
40,
41,
"metrics"
]
] | [
0,
1,
2,
3
] | [
[
4,
5,
28,
32,
"part-of",
"",
true,
false
],
[
40,
41,
35,
35,
"usage",
"",
true,
false
]
] | [
0,
1
] | [
"On",
"peut",
"utiliser",
"l'",
"algorithme",
"OSD",
"pour",
"dériver",
"des",
"limites",
"de",
"regret",
"mathématique",
"O",
"(",
"\\",
"sqrt",
"{T}",
")",
"/",
"math",
"pour",
"la",
"version",
"en",
"ligne",
"de",
"la",
"machine",
"à",
"vecteurs",
"de... | On peut utiliser l'algorithme OSD pour dériver des limites de regret mathématique O (\ sqrt {T}) / math pour la version en ligne de la machine à vecteurs de support pour la classification, qui utilise la perte charnière math v _t (w) =\ max\ {0, 1 - y _t (w\ cdot x _t)\} / math | [
[
0,
2
],
[
3,
7
],
[
8,
16
],
[
17,
19
],
[
19,
29
],
[
30,
33
],
[
34,
38
],
[
39,
46
],
[
47,
50
],
[
51,
58
],
[
59,
61
],
[
62,
68
],
[
69,
81
],
[
82,... |
ai-train-22 | [
[
4,
6,
"task"
],
[
9,
13,
"task"
],
[
12,
12,
"task"
],
[
16,
18,
"task"
],
[
21,
22,
"task"
],
[
25,
27,
"task"
],
[
30,
31,
"task"
],
[
34,
38,
"task"
],
[
41,
43,
"task"
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [] | [] | [
"Les",
"applications",
"comprennent",
"la",
"reconnaissance",
"d'",
"objets",
",",
"la",
"cartographie",
"et",
"la",
"navigation",
"robotiques",
",",
"l'",
"assemblage",
"d'",
"images",
",",
"la",
"modélisation",
"3D",
",",
"la",
"reconnaissance",
"de",
"gestes"... | Les applications comprennent la reconnaissance d'objets, la cartographie et la navigation robotiques, l'assemblage d'images, la modélisation 3D, la reconnaissance de gestes, le suivi vidéo, l'identification individuelle d'animaux sauvages et le déplacement d'allumettes. | [
[
0,
3
],
[
4,
16
],
[
17,
28
],
[
29,
31
],
[
32,
46
],
[
47,
49
],
[
49,
55
],
[
55,
56
],
[
57,
59
],
[
60,
72
],
[
73,
75
],
[
76,
78
],
[
79,
89
],
[
9... |
ai-train-23 | [
[
13,
16,
"task"
],
[
23,
24,
"university"
],
[
28,
30,
"university"
],
[
33,
34,
"university"
],
[
38,
39,
"university"
],
[
43,
48,
"university"
],
[
52,
54,
"university"
],
[
58,
60,
"unive... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11
] | [
[
13,
16,
23,
24,
"related-to",
"",
true,
false
],
[
13,
16,
28,
30,
"related-to",
"",
true,
false
],
[
13,
16,
33,
34,
"related-to",
"",
true,
false
],
[
13,
16,
38,
39,
"related-to",
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10
] | [
"Un",
"certain",
"nombre",
"de",
"groupes",
"et",
"d'",
"entreprises",
"effectuent",
"des",
"recherches",
"sur",
"l'",
"estimation",
"de",
"la",
"pose",
",",
"notamment",
"des",
"groupes",
"de",
"l'",
"université",
"Brown",
",",
"de",
"l'",
"université",
"Ca... | Un certain nombre de groupes et d'entreprises effectuent des recherches sur l'estimation de la pose, notamment des groupes de l'université Brown, de l'université Carnegie Mellon, du MPI Saarbruecken, de l'université Stanford, de l'université de Californie, San Diego, de l'université de Toronto, de l'École Centrale Pari... | [
[
0,
2
],
[
3,
10
],
[
11,
17
],
[
18,
20
],
[
21,
28
],
[
29,
31
],
[
32,
34
],
[
34,
45
],
[
46,
56
],
[
57,
60
],
[
61,
71
],
[
72,
75
],
[
76,
78
],
[
7... |
ai-train-24 | [
[
0,
6,
"metrics"
]
] | [
0
] | [] | [] | [
"La",
"fonction",
"Sigmoïde",
"Perte",
"d'",
"entropie",
"croisée",
"est",
"utilisée",
"pour",
"prédire",
"K",
"valeurs",
"de",
"probabilité",
"indépendantes",
"en",
"maths",
"0,1",
"/",
"maths",
"."
] | La fonction Sigmoïde Perte d'entropie croisée est utilisée pour prédire K valeurs de probabilité indépendantes en maths 0,1 / maths. | [
[
0,
2
],
[
3,
11
],
[
12,
20
],
[
21,
26
],
[
27,
29
],
[
29,
37
],
[
38,
45
],
[
46,
49
],
[
50,
58
],
[
59,
63
],
[
64,
71
],
[
72,
73
],
[
74,
81
],
[
8... |
ai-train-25 | [
[
6,
8,
"misc"
],
[
10,
10,
"field"
],
[
13,
13,
"field"
],
[
16,
18,
"university"
],
[
21,
21,
"country"
],
[
26,
28,
"misc"
],
[
31,
35,
"university"
],
[
38,
38,
"country"
],
[
45,
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
[
6,
8,
10,
10,
"topic",
"",
false,
false
],
[
6,
8,
13,
13,
"topic",
"",
false,
false
],
[
6,
8,
16,
18,
"physical",
"",
true,
false
],
[
16,
18,
21,
21,
"physical",
"",
false,
... | [
0,
1,
2,
3,
4,
5
] | [
"Il",
"a",
"été",
"titulaire",
"de",
"la",
"chaire",
"Johann",
"Bernoulli",
"de",
"mathématiques",
"et",
"d'",
"informatique",
"à",
"l'",
"université",
"de",
"Groningue",
",",
"aux",
"Pays-Bas",
",",
"et",
"de",
"la",
"chaire",
"Toshiba",
"Endowed",
"à",
"... | Il a été titulaire de la chaire Johann Bernoulli de mathématiques et d'informatique à l'université de Groningue, aux Pays-Bas, et de la chaire Toshiba Endowed à l'institut de technologie de Tokyo, au Japon, avant de devenir professeur à Cambridge. | [
[
0,
2
],
[
3,
4
],
[
5,
8
],
[
9,
18
],
[
19,
21
],
[
22,
24
],
[
25,
31
],
[
32,
38
],
[
39,
48
],
[
49,
51
],
[
52,
65
],
[
66,
68
],
[
69,
71
],
[
71,
... |
ai-train-26 | [
[
7,
9,
"algorithm"
],
[
14,
18,
"algorithm"
],
[
20,
20,
"algorithm"
],
[
25,
26,
"researcher"
],
[
28,
29,
"researcher"
]
] | [
0,
1,
2,
3,
4
] | [
[
7,
9,
14,
18,
"usage",
"",
true,
false
],
[
14,
18,
25,
26,
"origin",
"",
false,
false
],
[
14,
18,
28,
29,
"origin",
"",
false,
false
],
[
20,
20,
14,
18,
"named",
"",
false,
... | [
0,
1,
2,
3
] | [
"Une",
"autre",
"technique",
"particulièrement",
"utilisée",
"pour",
"les",
"réseaux",
"neuronaux",
"récurrents",
"est",
"le",
"réseau",
"à",
"mémoire",
"à",
"long",
"court",
"terme",
"(",
"LSTM",
")",
"de",
"1997",
"par",
"Sepp",
"Hochreiter",
"&",
"Jürgen",
... | Une autre technique particulièrement utilisée pour les réseaux neuronaux récurrents est le réseau à mémoire à long court terme (LSTM) de 1997 par Sepp Hochreiter & Jürgen Schmidhuber. | [
[
0,
3
],
[
4,
9
],
[
10,
19
],
[
20,
36
],
[
37,
45
],
[
46,
50
],
[
51,
54
],
[
55,
62
],
[
63,
72
],
[
73,
83
],
[
84,
87
],
[
88,
90
],
[
91,
97
],
[
98... |
ai-train-27 | [
[
5,
7,
"programlang"
],
[
9,
9,
"product"
],
[
16,
16,
"product"
],
[
49,
49,
"product"
]
] | [
0,
1,
2,
3
] | [
[
9,
9,
5,
7,
"general-affiliation",
"",
false,
false
],
[
9,
9,
16,
16,
"named",
"",
false,
false
]
] | [
0,
1
] | [
"L'",
"inclusion",
"d'",
"un",
"interpréteur",
"C",
"+",
"+",
"(",
"CINT",
"jusqu'",
"à",
"la",
"version",
"5.34",
",",
"Cling",
"à",
"partir",
"de",
"la",
"version",
"6",
")",
"rend",
"ce",
"paquet",
"très",
"polyvalent",
"car",
"il",
"peut",
"être",
... | L'inclusion d'un interpréteur C + + (CINT jusqu'à la version 5.34, Cling à partir de la version 6) rend ce paquet très polyvalent car il peut être utilisé en mode interactif, scripté et compilé d'une manière similaire aux produits commerciaux comme MATLAB. | [
[
0,
2
],
[
2,
11
],
[
12,
14
],
[
14,
16
],
[
17,
29
],
[
30,
31
],
[
32,
33
],
[
34,
35
],
[
36,
37
],
[
37,
41
],
[
42,
48
],
[
48,
49
],
[
50,
52
],
[
5... |
ai-train-28 | [
[
0,
2,
"product"
],
[
29,
32,
"field"
],
[
36,
38,
"task"
],
[
41,
44,
"task"
],
[
47,
49,
"task"
],
[
52,
54,
"task"
]
] | [
0,
1,
2,
3,
4,
5
] | [
[
0,
2,
29,
32,
"related-to",
"",
false,
false
],
[
36,
38,
29,
32,
"part-of",
"",
false,
false
],
[
41,
44,
29,
32,
"part-of",
"",
false,
false
],
[
47,
49,
29,
32,
"part-of",
"",
... | [
0,
1,
2,
3,
4
] | [
"Les",
"interfaces",
"vocales",
"qui",
"interprètent",
"et",
"gèrent",
"l'",
"état",
"de",
"la",
"conversation",
"sont",
"difficiles",
"à",
"concevoir",
"en",
"raison",
"de",
"la",
"difficulté",
"inhérente",
"à",
"l'",
"intégration",
"de",
"tâches",
"complexes",... | Les interfaces vocales qui interprètent et gèrent l'état de la conversation sont difficiles à concevoir en raison de la difficulté inhérente à l'intégration de tâches complexes de traitement du langage naturel telles que la résolution de coréférences, la reconnaissance d'entités nommées, la recherche d'informations et ... | [
[
0,
3
],
[
4,
14
],
[
15,
22
],
[
23,
26
],
[
27,
39
],
[
40,
42
],
[
43,
49
],
[
50,
52
],
[
52,
56
],
[
57,
59
],
[
60,
62
],
[
63,
75
],
[
76,
80
],
[
8... |
ai-train-29 | [
[
6,
8,
"algorithm"
],
[
11,
14,
"algorithm"
],
[
22,
23,
"researcher"
],
[
25,
28,
"organisation"
],
[
35,
37,
"field"
],
[
40,
41,
"field"
]
] | [
0,
1,
2,
3,
4,
5
] | [
[
6,
8,
22,
23,
"origin",
"",
false,
false
],
[
6,
8,
35,
37,
"part-of",
"",
false,
false
],
[
6,
8,
40,
41,
"part-of",
"",
false,
false
],
[
11,
14,
22,
23,
"origin",
"",
false,
... | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
"Entre",
"2009",
"et",
"2012",
",",
"les",
"réseaux",
"neuronaux",
"récurrents",
"et",
"les",
"réseaux",
"neuronaux",
"feedforward",
"profonds",
"développés",
"par",
"le",
"groupe",
"de",
"recherche",
"de",
"Jürgen",
"Schmidhuber",
"au",
"Swiss",
"AI",
"Lab",
... | Entre 2009 et 2012, les réseaux neuronaux récurrents et les réseaux neuronaux feedforward profonds développés par le groupe de recherche de Jürgen Schmidhuber au Swiss AI Lab IDSIA ont remporté huit concours internationaux de reconnaissance de formes et d'apprentissage automatique. | [
[
0,
5
],
[
6,
10
],
[
11,
13
],
[
14,
18
],
[
18,
19
],
[
20,
23
],
[
24,
31
],
[
32,
41
],
[
42,
52
],
[
53,
55
],
[
56,
59
],
[
60,
67
],
[
68,
77
],
[
7... |
ai-train-30 | [
[
0,
4,
"product"
],
[
10,
11,
"product"
],
[
13,
14,
"product"
],
[
20,
21,
"task"
],
[
24,
24,
"task"
]
] | [
0,
1,
2,
3,
4
] | [
[
0,
4,
10,
11,
"usage",
"",
false,
false
],
[
0,
4,
13,
14,
"usage",
"",
false,
false
],
[
0,
4,
20,
21,
"usage",
"",
true,
false
],
[
0,
4,
24,
24,
"usage",
"",
true,
false
... | [
0,
1,
2,
3
] | [
"Les",
"systèmes",
"de",
"bureau",
"Windows",
"modernes",
"peuvent",
"utiliser",
"les",
"composants",
"SAPI",
"4",
"et",
"SAPI",
"5",
"pour",
"prendre",
"en",
"charge",
"la",
"synthèse",
"vocale",
"et",
"la",
"parole",
"."
] | Les systèmes de bureau Windows modernes peuvent utiliser les composants SAPI 4 et SAPI 5 pour prendre en charge la synthèse vocale et la parole. | [
[
0,
3
],
[
4,
12
],
[
13,
15
],
[
16,
22
],
[
23,
30
],
[
31,
39
],
[
40,
47
],
[
48,
56
],
[
57,
60
],
[
61,
71
],
[
72,
76
],
[
77,
78
],
[
79,
81
],
[
8... |
ai-train-31 | [
[
8,
13,
"misc"
],
[
15,
15,
"field"
],
[
18,
20,
"university"
],
[
27,
30,
"field"
],
[
33,
37,
"university"
]
] | [
0,
1,
2,
3,
4
] | [
[
8,
13,
15,
15,
"topic",
"topic_of_award",
false,
false
],
[
8,
13,
18,
20,
"origin",
"",
true,
false
],
[
27,
30,
33,
37,
"related-to",
"",
true,
false
]
] | [
0,
1,
2
] | [
"Il",
"a",
"reçu",
"deux",
"diplômes",
"honorifiques",
",",
"un",
"S.",
"V.",
"della",
"laurea",
"ad",
"honorem",
"en",
"psychologie",
"de",
"l'",
"université",
"de",
"Padoue",
"en",
"1995",
"et",
"un",
"doctorat",
"en",
"conception",
"et",
"ingénierie",
"... | Il a reçu deux diplômes honorifiques, un S. V. della laurea ad honorem en psychologie de l'université de Padoue en 1995 et un doctorat en conception et ingénierie industrielles de l'université de technologie de Delft. | [
[
0,
2
],
[
3,
4
],
[
5,
9
],
[
10,
14
],
[
15,
23
],
[
24,
36
],
[
36,
37
],
[
38,
40
],
[
41,
43
],
[
44,
46
],
[
47,
52
],
[
53,
59
],
[
60,
62
],
[
63,
... |
ai-train-32 | [
[
7,
8,
"researcher"
],
[
13,
16,
"organisation"
],
[
18,
18,
"location"
],
[
20,
21,
"researcher"
],
[
33,
34,
"misc"
],
[
50,
52,
"misc"
],
[
72,
73,
"misc"
]
] | [
0,
1,
2,
3,
4,
5,
6
] | [
[
7,
8,
13,
16,
"physical",
"",
false,
false
],
[
7,
8,
13,
16,
"role",
"",
false,
false
],
[
13,
16,
18,
18,
"physical",
"",
false,
false
],
[
20,
21,
33,
34,
"related-to",
"works_wit... | [
0,
1,
2,
3,
4,
5
] | [
"Avec",
"son",
"collaborateur",
"de",
"longue",
"date",
",",
"Laurent",
"Cohen",
",",
"neurologue",
"à",
"l'",
"hôpital",
"de",
"la",
"Pitié-Salpêtrière",
"à",
"Paris",
",",
"Dehaene",
"a",
"également",
"identifié",
"des",
"patients",
"présentant",
"des",
"lés... | Avec son collaborateur de longue date, Laurent Cohen, neurologue à l'hôpital de la Pitié-Salpêtrière à Paris, Dehaene a également identifié des patients présentant des lésions dans différentes régions du lobe pariétal avec une multiplication altérée, mais une soustraction préservée (associée à des lésions du lobule par... | [
[
0,
4
],
[
5,
8
],
[
9,
22
],
[
23,
25
],
[
26,
32
],
[
33,
37
],
[
37,
38
],
[
39,
46
],
[
47,
52
],
[
52,
53
],
[
54,
64
],
[
65,
66
],
[
67,
69
],
[
69,... |
ai-train-33 | [
[
7,
9,
"product"
],
[
15,
17,
"misc"
],
[
19,
20,
"misc"
],
[
28,
28,
"misc"
]
] | [
0,
1,
2,
3
] | [
[
15,
17,
7,
9,
"topic",
"",
false,
false
],
[
19,
20,
7,
9,
"topic",
"",
false,
false
],
[
28,
28,
7,
9,
"topic",
"",
false,
false
]
] | [
0,
1,
2
] | [
"Plus",
"récemment",
",",
"les",
"représentations",
"fictionnelles",
"de",
"robots",
"artificiellement",
"intelligents",
"dans",
"des",
"films",
"tels",
"que",
"A.I.",
"Artificial",
"Intelligence",
"et",
"Ex",
"Machina",
"et",
"l'",
"adaptation",
"télévisée",
"de",
... | Plus récemment, les représentations fictionnelles de robots artificiellement intelligents dans des films tels que A.I. Artificial Intelligence et Ex Machina et l'adaptation télévisée de 2016 de Westworld ont suscité la sympathie du public pour les robots eux-mêmes. | [
[
0,
4
],
[
5,
14
],
[
14,
15
],
[
16,
19
],
[
20,
35
],
[
36,
49
],
[
50,
52
],
[
53,
59
],
[
60,
76
],
[
77,
89
],
[
90,
94
],
[
95,
98
],
[
99,
104
],
[
... |
ai-train-34 | [
[
7,
9,
"field"
],
[
12,
15,
"algorithm"
],
[
18,
20,
"task"
]
] | [
0,
1,
2
] | [
[
12,
15,
7,
9,
"part-of",
"",
false,
false
],
[
18,
20,
7,
9,
"part-of",
"",
false,
false
]
] | [
0,
1
] | [
"Deux",
"des",
"principales",
"méthodes",
"utilisées",
"dans",
"l'",
"apprentissage",
"non",
"supervisé",
"sont",
"l'",
"analyse",
"en",
"composantes",
"principales",
"et",
"l'",
"analyse",
"en",
"grappes",
"."
] | Deux des principales méthodes utilisées dans l'apprentissage non supervisé sont l'analyse en composantes principales et l'analyse en grappes. | [
[
0,
4
],
[
5,
8
],
[
9,
20
],
[
21,
29
],
[
30,
39
],
[
40,
44
],
[
45,
47
],
[
47,
60
],
[
61,
64
],
[
65,
74
],
[
75,
79
],
[
80,
82
],
[
82,
89
],
[
90,... |
ai-train-35 | [
[
0,
3,
"organisation"
],
[
27,
28,
"misc"
],
[
33,
34,
"misc"
],
[
36,
38,
"person"
],
[
43,
44,
"person"
]
] | [
0,
1,
2,
3,
4
] | [
[
27,
28,
0,
3,
"artifact",
"",
false,
false
],
[
33,
34,
0,
3,
"artifact",
"",
false,
false
],
[
33,
34,
36,
38,
"role",
"director_of",
false,
false
],
[
33,
34,
43,
44,
"role",
"acto... | [
0,
1,
2,
3
] | [
"La",
"Walt",
"Disney",
"Company",
"a",
"également",
"commencé",
"à",
"utiliser",
"de",
"manière",
"plus",
"importante",
"les",
"films",
"en",
"3D",
"dans",
"des",
"lieux",
"spéciaux",
"pour",
"impressionner",
"le",
"public",
".",
"Les",
"Voyages",
"magiques",... | La Walt Disney Company a également commencé à utiliser de manière plus importante les films en 3D dans des lieux spéciaux pour impressionner le public. Les Voyages magiques (1982) et Captain EO (Francis Ford Coppola, 1986, avec Michael Jackson) en sont des exemples notables. | [
[
0,
2
],
[
3,
7
],
[
8,
14
],
[
15,
22
],
[
23,
24
],
[
25,
34
],
[
35,
43
],
[
44,
45
],
[
46,
54
],
[
55,
57
],
[
58,
65
],
[
66,
70
],
[
71,
81
],
[
82,... |
ai-train-36 | [
[
14,
17,
"field"
],
[
24,
28,
"task"
],
[
31,
32,
"task"
],
[
34,
34,
"researcher"
]
] | [
0,
1,
2,
3
] | [
[
24,
28,
14,
17,
"part-of",
"",
false,
false
],
[
31,
32,
14,
17,
"part-of",
"",
false,
false
]
] | [
0,
1
] | [
"Depuis",
"2002",
",",
"l'",
"entraînement",
"par",
"perceptron",
"est",
"devenu",
"populaire",
"dans",
"le",
"domaine",
"du",
"traitement",
"du",
"langage",
"naturel",
"pour",
"des",
"tâches",
"telles",
"que",
"l'",
"étiquetage",
"des",
"parties",
"du",
"disc... | Depuis 2002, l'entraînement par perceptron est devenu populaire dans le domaine du traitement du langage naturel pour des tâches telles que l'étiquetage des parties du discours et l'analyse syntaxique (Collins, 2002). | [
[
0,
6
],
[
7,
11
],
[
11,
12
],
[
13,
15
],
[
15,
27
],
[
28,
31
],
[
32,
42
],
[
43,
46
],
[
47,
53
],
[
54,
63
],
[
64,
68
],
[
69,
71
],
[
72,
79
],
[
8... |
ai-train-37 | [
[
2,
4,
"product"
],
[
12,
16,
"organisation"
],
[
18,
19,
"organisation"
],
[
21,
21,
"country"
],
[
25,
30,
"product"
],
[
35,
36,
"researcher"
],
[
47,
47,
"organisation"
]
] | [
0,
1,
2,
3,
4,
5,
6
] | [
[
12,
16,
2,
4,
"role",
"introduces_to_market",
true,
false
],
[
18,
19,
2,
4,
"role",
"introduces_to_market",
true,
false
],
[
18,
19,
21,
21,
"physical",
"",
false,
false
],
[
25,
30,
47,
47... | [
0,
1,
2,
3,
4
] | [
"Le",
"premier",
"robot",
"de",
"palettisation",
"a",
"été",
"introduit",
"en",
"1963",
"par",
"la",
"société",
"Fuji",
"Yusoki",
"Kogyo",
".",
"par",
"KUKA",
"robotics",
"en",
"Allemagne",
",",
"et",
"la",
"machine",
"universelle",
"programmable",
"pour",
"... | Le premier robot de palettisation a été introduit en 1963 par la société Fuji Yusoki Kogyo. par KUKA robotics en Allemagne, et la machine universelle programmable pour l'assemblage a été inventée par Victor Scheinman en 1976, et la conception a été vendue à Unimation. | [
[
0,
2
],
[
3,
10
],
[
11,
16
],
[
17,
19
],
[
20,
33
],
[
34,
35
],
[
36,
39
],
[
40,
49
],
[
50,
52
],
[
53,
57
],
[
58,
61
],
[
62,
64
],
[
65,
72
],
[
7... |
ai-train-38 | [
[
13,
13,
"conference"
],
[
15,
16,
"researcher"
],
[
27,
27,
"field"
],
[
44,
45,
"researcher"
],
[
53,
54,
"researcher"
],
[
67,
67,
"field"
]
] | [
0,
1,
2,
3,
4,
5
] | [
[
15,
16,
13,
13,
"role",
"president_of",
false,
false
],
[
15,
16,
44,
45,
"role",
"colleagues",
false,
false
],
[
27,
27,
67,
67,
"named",
"same",
false,
false
]
] | [
0,
1,
2
] | [
"Au",
"milieu",
"des",
"années",
"1990",
",",
"alors",
"qu'",
"il",
"était",
"président",
"de",
"l'",
"AAAI",
",",
"Hayes",
"a",
"lancé",
"une",
"série",
"d'",
"attaques",
"contre",
"les",
"critiques",
"de",
"l'",
"IA",
",",
"le",
"plus",
"souvent",
"f... | Au milieu des années 1990, alors qu'il était président de l'AAAI, Hayes a lancé une série d'attaques contre les critiques de l'IA, le plus souvent formulées de manière ironique, et a inventé (avec son collègue Kenneth Ford) un prix portant le nom de Simon Newcomb et récompensant l'argument le plus ridicule réfutant la ... | [
[
0,
2
],
[
3,
9
],
[
10,
13
],
[
14,
20
],
[
21,
25
],
[
25,
26
],
[
27,
32
],
[
33,
36
],
[
36,
38
],
[
39,
44
],
[
45,
54
],
[
55,
57
],
[
58,
60
],
[
60... |
ai-train-39 | [
[
16,
19,
"algorithm"
],
[
46,
47,
"algorithm"
],
[
60,
65,
"algorithm"
],
[
69,
71,
"algorithm"
]
] | [
0,
1,
2,
3
] | [
[
16,
19,
46,
47,
"named",
"same",
false,
false
],
[
60,
65,
16,
19,
"type-of",
"",
false,
false
],
[
69,
71,
16,
19,
"type-of",
"",
false,
false
]
] | [
0,
1,
2
] | [
"Une",
"valeur",
"optimale",
"de",
"mathématique",
"alpha",
"/",
"math",
"peut",
"être",
"trouvée",
"à",
"l'",
"aide",
"d'",
"un",
"algorithme",
"de",
"recherche",
"linéaire",
",",
"c'est-à-dire",
"que",
"la",
"valeur",
"de",
"mathématique",
"alpha",
"/",
"m... | Une valeur optimale de mathématique alpha / math peut être trouvée à l'aide d'un algorithme de recherche linéaire, c'est-à-dire que la valeur de mathématique alpha / math est déterminée en trouvant la valeur qui minimise S, généralement à l'aide d'une recherche linéaire dans l'intervalle math0\ alpha 1 / math ou d'une ... | [
[
0,
3
],
[
4,
10
],
[
11,
19
],
[
20,
22
],
[
23,
35
],
[
36,
41
],
[
42,
43
],
[
44,
48
],
[
49,
53
],
[
54,
58
],
[
59,
66
],
[
67,
68
],
[
69,
71
],
[
7... |
ai-train-40 | [
[
5,
7,
"algorithm"
],
[
9,
10,
"algorithm"
],
[
20,
21,
"product"
]
] | [
0,
1,
2
] | [] | [] | [
"Il",
"aborde",
"les",
"techniques",
"de",
"recherche",
"en",
"largeur",
"et",
"en",
"profondeur",
",",
"mais",
"conclut",
"finalement",
"que",
"les",
"résultats",
"représentent",
"des",
"systèmes",
"experts",
"qui",
"incarnent",
"beaucoup",
"de",
"connaissances",... | Il aborde les techniques de recherche en largeur et en profondeur, mais conclut finalement que les résultats représentent des systèmes experts qui incarnent beaucoup de connaissances techniques, mais ne font pas beaucoup de lumière sur les processus mentaux que les humains utilisent pour résoudre de telles énigmes. | [
[
0,
2
],
[
3,
9
],
[
10,
13
],
[
14,
24
],
[
25,
27
],
[
28,
37
],
[
38,
40
],
[
41,
48
],
[
49,
51
],
[
52,
54
],
[
55,
65
],
[
65,
66
],
[
67,
71
],
[
72... |
ai-train-41 | [
[
0,
5,
"task"
],
[
4,
4,
"task"
]
] | [
0,
1
] | [] | [] | [
"La",
"reconnaissance",
"et",
"la",
"synthèse",
"vocales",
"traitent",
"de",
"la",
"manière",
"dont",
"le",
"langage",
"parlé",
"peut",
"être",
"compris",
"ou",
"créé",
"à",
"l'",
"aide",
"d'",
"ordinateurs",
"."
] | La reconnaissance et la synthèse vocales traitent de la manière dont le langage parlé peut être compris ou créé à l'aide d'ordinateurs. | [
[
0,
2
],
[
3,
17
],
[
18,
20
],
[
21,
23
],
[
24,
32
],
[
33,
40
],
[
41,
49
],
[
50,
52
],
[
53,
55
],
[
56,
63
],
[
64,
68
],
[
69,
71
],
[
72,
79
],
[
8... |
ai-train-42 | [
[
18,
20,
"algorithm"
],
[
35,
37,
"algorithm"
]
] | [
0,
1
] | [] | [] | [
"Cette",
"mathématique",
"du",
"thêta",
"^",
"{*}",
"/",
"math",
"est",
"normalement",
"estimée",
"à",
"l'",
"aide",
"d'",
"une",
"procédure",
"de",
"Maximum",
"de",
"Vraisemblance",
"(",
"math\\\\_theta",
"^",
"{*}",
"=",
"\\",
"theta",
"^",
"{ML}",
"/",
... | Cette mathématique du thêta ^ {*} / math est normalement estimée à l'aide d'une procédure de Maximum de Vraisemblance (math\\_theta ^ {*} =\theta ^ {ML} / math) ou de Maximum A Posteriori (math\_theta ^ {*} =\theta ^ {MAP} / math). | [
[
0,
5
],
[
6,
18
],
[
19,
21
],
[
22,
27
],
[
28,
29
],
[
30,
33
],
[
34,
35
],
[
36,
40
],
[
41,
44
],
[
45,
56
],
[
57,
64
],
[
65,
66
],
[
67,
69
],
[
6... |
ai-train-43 | [
[
6,
8,
"product"
]
] | [
0
] | [] | [] | [
"Certaines",
"langues",
"moins",
"répandues",
"utilisent",
"le",
"synthétiseur",
"open-source",
"eSpeak",
"pour",
"leur",
"discours",
",",
"produisant",
"une",
"voix",
"robotique",
"et",
"maladroite",
"qui",
"peut",
"être",
"difficile",
"à",
"comprendre",
"."
] | Certaines langues moins répandues utilisent le synthétiseur open-source eSpeak pour leur discours, produisant une voix robotique et maladroite qui peut être difficile à comprendre. | [
[
0,
9
],
[
10,
17
],
[
18,
23
],
[
24,
33
],
[
34,
43
],
[
44,
46
],
[
47,
59
],
[
60,
71
],
[
72,
78
],
[
79,
83
],
[
84,
88
],
[
89,
97
],
[
97,
98
],
[
... |
ai-train-44 | [
[
25,
25,
"programlang"
],
[
47,
48,
"programlang"
],
[
50,
50,
"product"
]
] | [
0,
1,
2
] | [
[
25,
25,
47,
48,
"compare",
"",
false,
false
],
[
25,
25,
50,
50,
"compare",
"",
false,
false
]
] | [
0,
1
] | [
"Bien",
"qu'",
"il",
"soit",
"principalement",
"utilisé",
"par",
"les",
"statisticiens",
"et",
"autres",
"praticiens",
"nécessitant",
"un",
"environnement",
"pour",
"le",
"calcul",
"statistique",
"et",
"le",
"développement",
"de",
"logiciels",
",",
"R",
"peut",
... | Bien qu'il soit principalement utilisé par les statisticiens et autres praticiens nécessitant un environnement pour le calcul statistique et le développement de logiciels, R peut également fonctionner comme une boîte à outils générale pour le calcul matriciel - avec des performances comparables à celles de GNU Octave o... | [
[
0,
4
],
[
5,
8
],
[
8,
10
],
[
11,
15
],
[
16,
30
],
[
31,
38
],
[
39,
42
],
[
43,
46
],
[
47,
60
],
[
61,
63
],
[
64,
70
],
[
71,
81
],
[
82,
93
],
[
94,... |
ai-train-45 | [
[
0,
1,
"algorithm"
],
[
6,
8,
"field"
],
[
12,
13,
"misc"
],
[
14,
15,
"researcher"
]
] | [
0,
1,
2,
3
] | [
[
0,
1,
6,
8,
"part-of",
"",
false,
false
],
[
0,
1,
14,
15,
"origin",
"",
false,
false
],
[
12,
13,
14,
15,
"named",
"",
false,
false
]
] | [
0,
1,
2
] | [
"L'",
"hétérodynage",
"est",
"une",
"technique",
"de",
"traitement",
"du",
"signal",
"inventée",
"par",
"l'",
"ingénieur-inventeur",
"canadien",
"Reginald",
"Fessenden",
"qui",
"permet",
"de",
"créer",
"de",
"nouvelles",
"fréquences",
"en",
"mélangeant",
"deux",
"... | L'hétérodynage est une technique de traitement du signal inventée par l'ingénieur-inventeur canadien Reginald Fessenden qui permet de créer de nouvelles fréquences en mélangeant deux fréquences. | [
[
0,
2
],
[
2,
14
],
[
15,
18
],
[
19,
22
],
[
23,
32
],
[
33,
35
],
[
36,
46
],
[
47,
49
],
[
50,
56
],
[
57,
65
],
[
66,
69
],
[
70,
72
],
[
72,
91
],
[
9... |
ai-train-46 | [
[
21,
22,
"person"
],
[
16,
18,
"misc"
],
[
26,
27,
"organisation"
],
[
34,
34,
"organisation"
],
[
30,
32,
"misc"
],
[
36,
37,
"person"
],
[
44,
44,
"organisation"
],
[
40,
42,
"misc"
],
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9
] | [
[
21,
22,
16,
18,
"role",
"actor_in",
false,
false
],
[
16,
18,
26,
27,
"artifact",
"",
false,
false
],
[
30,
32,
34,
34,
"artifact",
"",
false,
false
],
[
36,
37,
30,
32,
"role",
"act... | [
0,
1,
2,
3,
4,
5,
6
] | [
"Plusieurs",
"autres",
"films",
"ont",
"contribué",
"à",
"remettre",
"la",
"3D",
"sur",
"la",
"carte",
"ce",
"mois",
"-là",
":",
"Hondo",
",",
"un",
"film",
"de",
"John",
"Wayne",
"(",
"distribué",
"par",
"Warner",
"Bros.",
")",
",",
"Miss",
"Sadie",
"... | Plusieurs autres films ont contribué à remettre la 3D sur la carte ce mois-là : Hondo, un film de John Wayne (distribué par Warner Bros.), Miss Sadie Thompson de Columbia avec Rita Hayworth, et Money From Home de Paramount avec Dean Martin et Jerry Lewis. | [
[
0,
9
],
[
10,
16
],
[
17,
22
],
[
23,
26
],
[
27,
36
],
[
37,
38
],
[
39,
47
],
[
48,
50
],
[
51,
53
],
[
54,
57
],
[
58,
60
],
[
61,
66
],
[
67,
69
],
[
... |
ai-train-47 | [
[
0,
0,
"product"
],
[
8,
9,
"field"
],
[
5,
7,
"task"
],
[
17,
17,
"organisation"
]
] | [
0,
1,
2,
3
] | [
[
0,
0,
5,
7,
"general-affiliation",
"",
false,
false
],
[
0,
0,
17,
17,
"artifact",
"",
false,
false
],
[
5,
7,
8,
9,
"part-of",
"task_part_of_field",
false,
false
]
] | [
0,
1,
2
] | [
"DeepFace",
"est",
"un",
"système",
"de",
"reconnaissance",
"faciale",
"par",
"apprentissage",
"profond",
"créé",
"par",
"un",
"groupe",
"de",
"recherche",
"de",
"Facebook",
"."
] | DeepFace est un système de reconnaissance faciale par apprentissage profond créé par un groupe de recherche de Facebook. | [
[
0,
8
],
[
9,
12
],
[
13,
15
],
[
16,
23
],
[
24,
26
],
[
27,
41
],
[
42,
49
],
[
50,
53
],
[
54,
67
],
[
68,
75
],
[
76,
80
],
[
81,
84
],
[
85,
87
],
[
8... |
ai-train-48 | [
[
1,
4,
"field"
],
[
12,
12,
"conference"
],
[
20,
20,
"field"
],
[
27,
34,
"conference"
]
] | [
0,
1,
2,
3
] | [
[
1,
4,
20,
20,
"part-of",
"subfield",
false,
false
],
[
12,
12,
1,
4,
"topic",
"",
false,
false
],
[
27,
34,
1,
4,
"topic",
"",
false,
false
]
] | [
0,
1,
2
] | [
"Le",
"traitement",
"de",
"la",
"géométrie",
"est",
"un",
"sujet",
"de",
"recherche",
"courant",
"au",
"SIGGRAPH",
",",
"la",
"principale",
"conférence",
"universitaire",
"sur",
"l'",
"infographie",
",",
"et",
"le",
"sujet",
"principal",
"du",
"Symposium",
"an... | Le traitement de la géométrie est un sujet de recherche courant au SIGGRAPH, la principale conférence universitaire sur l'infographie, et le sujet principal du Symposium annuel sur le traitement de la géométrie. | [
[
0,
2
],
[
3,
13
],
[
14,
16
],
[
17,
19
],
[
20,
29
],
[
30,
33
],
[
34,
36
],
[
37,
42
],
[
43,
45
],
[
46,
55
],
[
56,
63
],
[
64,
66
],
[
67,
75
],
[
7... |
ai-train-49 | [
[
0,
3,
"task"
],
[
6,
8,
"task"
],
[
21,
24,
"algorithm"
],
[
26,
26,
"algorithm"
],
[
30,
32,
"algorithm"
],
[
34,
34,
"algorithm"
],
[
38,
41,
"algorithm"
],
[
43,
43,
"algorithm"
],
[
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9
] | [
[
21,
24,
48,
48,
"general-affiliation",
"",
false,
false
],
[
26,
26,
21,
24,
"named",
"",
false,
false
],
[
30,
32,
48,
48,
"general-affiliation",
"",
false,
false
],
[
34,
34,
30,
32,
"... | [
0,
1,
2,
3,
4,
5
] | [
"L'",
"extraction",
"de",
"caractéristiques",
"et",
"la",
"réduction",
"de",
"dimension",
"peuvent",
"être",
"combinées",
"en",
"une",
"seule",
"étape",
"en",
"utilisant",
"les",
"techniques",
"d'",
"analyse",
"en",
"composantes",
"principales",
"(",
"ACP",
")",... | L'extraction de caractéristiques et la réduction de dimension peuvent être combinées en une seule étape en utilisant les techniques d'analyse en composantes principales (ACP), d'analyse discriminante linéaire (ADL) ou d'analyse de corrélation canonique (ACC) comme étape de prétraitement, suivie d'un regroupement par k ... | [
[
0,
2
],
[
2,
12
],
[
13,
15
],
[
16,
32
],
[
33,
35
],
[
36,
38
],
[
39,
48
],
[
49,
51
],
[
52,
61
],
[
62,
69
],
[
70,
74
],
[
75,
84
],
[
85,
87
],
[
8... |
ai-train-50 | [
[
0,
3,
"algorithm"
],
[
12,
13,
"field"
],
[
16,
18,
"field"
]
] | [
0,
1,
2
] | [
[
0,
3,
12,
13,
"related-to",
"good_at",
true,
false
],
[
0,
3,
16,
18,
"related-to",
"good_at",
true,
false
]
] | [
0,
1
] | [
"Les",
"réseaux",
"neuronaux",
"artificiels",
"sont",
"des",
"modèles",
"informatiques",
"qui",
"excellent",
"dans",
"l'",
"apprentissage",
"automatique",
"et",
"la",
"reconnaissance",
"des",
"formes",
"."
] | Les réseaux neuronaux artificiels sont des modèles informatiques qui excellent dans l'apprentissage automatique et la reconnaissance des formes. | [
[
0,
3
],
[
4,
11
],
[
12,
21
],
[
22,
33
],
[
34,
38
],
[
39,
42
],
[
43,
50
],
[
51,
64
],
[
65,
68
],
[
69,
78
],
[
79,
83
],
[
84,
86
],
[
86,
99
],
[
1... |
ai-train-51 | [
[
0,
1,
"researcher"
],
[
3,
4,
"researcher"
],
[
6,
10,
"misc"
],
[
12,
16,
"conference"
],
[
18,
18,
"conference"
],
[
35,
38,
"algorithm"
],
[
39,
40,
"researcher"
],
[
42,
43,
"researcher"... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10
] | [
[
6,
10,
0,
1,
"artifact",
"",
false,
false
],
[
6,
10,
3,
4,
"artifact",
"",
false,
false
],
[
6,
10,
12,
16,
"temporal",
"",
false,
false
],
[
18,
18,
12,
16,
"named",
"",
false,... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
"C.",
"Papageorgiou",
"et",
"T.",
"Poggio",
",",
"A",
"Trainable",
"Pedestrian",
"Detection",
"system",
",",
"International",
"Journal",
"of",
"Computer",
"Vision",
"(",
"IJCV",
")",
",",
"pages",
"1",
":",
"15-33",
",",
"2000",
"d'",
"autres",
"utilisent",
... | C. Papageorgiou et T. Poggio, A Trainable Pedestrian Detection system, International Journal of Computer Vision (IJCV), pages 1 : 15-33, 2000 d'autres utilisent des caractéristiques locales comme l'histogramme des gradients orientés N. Dalal, B. Triggs, Histograms of oriented gradients for human detection, IEEE Compute... | [
[
0,
2
],
[
3,
15
],
[
16,
18
],
[
19,
21
],
[
22,
28
],
[
28,
29
],
[
30,
31
],
[
32,
41
],
[
42,
52
],
[
53,
62
],
[
63,
69
],
[
69,
70
],
[
71,
84
],
[
8... |
ai-train-52 | [
[
1,
1,
"algorithm"
],
[
6,
8,
"algorithm"
],
[
13,
15,
"task"
],
[
18,
19,
"field"
]
] | [
0,
1,
2,
3
] | [
[
1,
1,
6,
8,
"type-of",
"",
false,
false
],
[
13,
15,
1,
1,
"usage",
"",
true,
false
],
[
13,
15,
18,
19,
"part-of",
"task_part_of_field",
false,
false
]
] | [
0,
1,
2
] | [
"Un",
"autoencodeur",
"est",
"un",
"type",
"de",
"réseau",
"neuronal",
"artificiel",
"utilisé",
"pour",
"apprendre",
"l'",
"apprentissage",
"des",
"fonctionnalités",
"de",
"manière",
"non",
"supervisée",
"."
] | Un autoencodeur est un type de réseau neuronal artificiel utilisé pour apprendre l'apprentissage des fonctionnalités de manière non supervisée. | [
[
0,
2
],
[
3,
15
],
[
16,
19
],
[
20,
22
],
[
23,
27
],
[
28,
30
],
[
31,
37
],
[
38,
46
],
[
47,
57
],
[
58,
65
],
[
66,
70
],
[
71,
80
],
[
81,
83
],
[
8... |
ai-train-53 | [
[
0,
1,
"researcher"
],
[
6,
6,
"organisation"
],
[
15,
17,
"field"
],
[
20,
22,
"field"
],
[
27,
31,
"organisation"
],
[
33,
33,
"organisation"
],
[
43,
45,
"field"
],
[
48,
50,
"field"
],
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
[
0,
1,
6,
6,
"role",
"fellow_of",
false,
false
],
[
0,
1,
15,
17,
"related-to",
"contributes_to",
false,
false
],
[
0,
1,
20,
22,
"related-to",
"contributes_to",
false,
false
],
[
0,
1,
27,
3... | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
"M.",
"Haralick",
"est",
"membre",
"de",
"l'",
"IEEE",
"pour",
"ses",
"contributions",
"dans",
"le",
"domaine",
"de",
"la",
"vision",
"par",
"ordinateur",
"et",
"du",
"traitement",
"des",
"images",
"et",
"membre",
"de",
"l'",
"International",
"Association",
... | M. Haralick est membre de l'IEEE pour ses contributions dans le domaine de la vision par ordinateur et du traitement des images et membre de l'International Association for Pattern Recognition (IAPR) pour ses contributions dans le domaine de la reconnaissance des formes, du traitement des images et pour les services re... | [
[
0,
2
],
[
3,
11
],
[
12,
15
],
[
16,
22
],
[
23,
25
],
[
26,
28
],
[
28,
32
],
[
33,
37
],
[
38,
41
],
[
42,
55
],
[
56,
60
],
[
61,
63
],
[
64,
71
],
[
7... |
ai-train-54 | [
[
4,
8,
"task"
],
[
18,
20,
"algorithm"
],
[
22,
22,
"algorithm"
],
[
26,
27,
"researcher"
],
[
29,
30,
"organisation"
],
[
32,
33,
"researcher"
],
[
36,
38,
"university"
]
] | [
0,
1,
2,
3,
4,
5,
6
] | [
[
4,
8,
18,
20,
"usage",
"",
false,
false
],
[
18,
20,
26,
27,
"origin",
"",
true,
false
],
[
18,
20,
32,
33,
"origin",
"",
true,
false
],
[
22,
22,
18,
20,
"named",
"",
false,
... | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
"La",
"première",
"tentative",
"de",
"RAS",
"de",
"bout",
"en",
"bout",
"a",
"été",
"réalisée",
"avec",
"des",
"systèmes",
"basés",
"sur",
"la",
"classification",
"temporelle",
"connexionniste",
"(",
"CTC",
")",
"présentés",
"par",
"Alex",
"Graves",
"de",
"G... | La première tentative de RAS de bout en bout a été réalisée avec des systèmes basés sur la classification temporelle connexionniste (CTC) présentés par Alex Graves de Google DeepMind et Navdeep Jaitly de l'Université de Toronto en 2014. | [
[
0,
2
],
[
3,
11
],
[
12,
21
],
[
22,
24
],
[
25,
28
],
[
29,
31
],
[
32,
36
],
[
37,
39
],
[
40,
44
],
[
45,
46
],
[
47,
50
],
[
51,
59
],
[
60,
64
],
[
6... |
ai-train-55 | [
[
0,
2,
"algorithm"
],
[
4,
4,
"algorithm"
],
[
11,
12,
"algorithm"
],
[
14,
14,
"algorithm"
]
] | [
0,
1,
2,
3
] | [
[
4,
4,
0,
2,
"named",
"",
false,
false
],
[
11,
12,
0,
2,
"type-of",
"",
false,
false
],
[
14,
14,
11,
12,
"named",
"",
false,
false
]
] | [
0,
1,
2
] | [
"La",
"programmation",
"linéaire-fractionnelle",
"(",
"LFP",
")",
"est",
"une",
"généralisation",
"de",
"la",
"programmation",
"linéaire",
"(",
"LP",
")",
"."
] | La programmation linéaire-fractionnelle (LFP) est une généralisation de la programmation linéaire (LP). | [
[
0,
2
],
[
3,
16
],
[
17,
39
],
[
40,
41
],
[
41,
44
],
[
44,
45
],
[
46,
49
],
[
50,
53
],
[
54,
68
],
[
69,
71
],
[
72,
74
],
[
75,
88
],
[
89,
97
],
[
9... |
ai-train-56 | [
[
0,
1,
"researcher"
],
[
10,
11,
"misc"
],
[
14,
21,
"conference"
]
] | [
0,
1,
2
] | [
[
0,
1,
10,
11,
"win-defeat",
"",
false,
false
],
[
10,
11,
14,
21,
"temporal",
"",
false,
false
]
] | [
0,
1
] | [
"M.",
"Lafferty",
"a",
"reçu",
"de",
"nombreux",
"prix",
",",
"dont",
"deux",
"prix",
"Test-of-Time",
"à",
"l'",
"International",
"Conference",
"on",
"Machine",
"Learning",
"2011",
"&",
"2012",
","
] | M. Lafferty a reçu de nombreux prix, dont deux prix Test-of-Time à l'International Conference on Machine Learning 2011 & 2012, | [
[
0,
2
],
[
3,
11
],
[
12,
13
],
[
14,
18
],
[
19,
21
],
[
22,
30
],
[
31,
35
],
[
35,
36
],
[
37,
41
],
[
42,
46
],
[
47,
51
],
[
52,
64
],
[
65,
66
],
[
6... |
ai-train-57 | [
[
11,
11,
"product"
],
[
13,
13,
"programlang"
],
[
28,
29,
"algorithm"
]
] | [
0,
1,
2
] | [] | [] | [
"Avec",
"l'",
"avènement",
"des",
"cadres",
"basés",
"sur",
"des",
"composants",
"tels",
"que",
".NET",
"et",
"Java",
",",
"les",
"environnements",
"de",
"développement",
"basés",
"sur",
"des",
"composants",
"sont",
"capables",
"de",
"déployer",
"le",
"réseau"... | Avec l'avènement des cadres basés sur des composants tels que .NET et Java, les environnements de développement basés sur des composants sont capables de déployer le réseau neuronal développé dans ces cadres en tant que composants héritables. | [
[
0,
4
],
[
5,
7
],
[
7,
16
],
[
17,
20
],
[
21,
27
],
[
28,
33
],
[
34,
37
],
[
38,
41
],
[
42,
52
],
[
53,
57
],
[
58,
61
],
[
62,
66
],
[
67,
69
],
[
70,... |
ai-train-58 | [
[
3,
3,
"metrics"
]
] | [
0
] | [] | [] | [
"Comme",
"pour",
"le",
"BLEU",
",",
"l'",
"unité",
"de",
"base",
"de",
"l'",
"évaluation",
"est",
"la",
"phrase",
".",
"L'",
"algorithme",
"crée",
"d'",
"abord",
"un",
"alignement",
"(",
"voir",
"illustrations",
")",
"entre",
"deux",
"phrases",
",",
"la"... | Comme pour le BLEU, l'unité de base de l'évaluation est la phrase. L'algorithme crée d'abord un alignement (voir illustrations) entre deux phrases, la chaîne de traduction candidate et la chaîne de traduction de référence. | [
[
0,
5
],
[
6,
10
],
[
11,
13
],
[
14,
18
],
[
18,
19
],
[
20,
22
],
[
22,
27
],
[
28,
30
],
[
31,
35
],
[
36,
38
],
[
39,
41
],
[
41,
51
],
[
52,
55
],
[
5... |
ai-train-59 | [
[
6,
15,
"conference"
],
[
30,
30,
"task"
],
[
28,
33,
"task"
],
[
37,
38,
"metrics"
],
[
40,
44,
"metrics"
],
[
49,
52,
"conference"
],
[
54,
54,
"conference"
],
[
57,
57,
"location"
],
[... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
[
6,
15,
30,
30,
"related-to",
"subject_at",
false,
false
],
[
6,
15,
28,
33,
"related-to",
"subject_at",
false,
false
],
[
37,
38,
6,
15,
"temporal",
"",
false,
false
],
[
40,
44,
37,
38,
... | [
0,
1,
2,
3,
4,
5
] | [
"L'",
"une",
"des",
"métriques",
"utilisées",
"lors",
"des",
"conférences",
"annuelles",
"du",
"NIST",
"sur",
"la",
"compréhension",
"des",
"documents",
",",
"dans",
"lesquelles",
"des",
"groupes",
"de",
"recherche",
"soumettent",
"leurs",
"systèmes",
"pour",
"d... | L'une des métriques utilisées lors des conférences annuelles du NIST sur la compréhension des documents, dans lesquelles des groupes de recherche soumettent leurs systèmes pour des tâches de résumé et de traduction, est la métrique ROUGE (Recall-Oriented Understudy for Gisting Evaluation, In Advances of Neural Informat... | [
[
0,
2
],
[
2,
5
],
[
6,
9
],
[
10,
19
],
[
20,
29
],
[
30,
34
],
[
35,
38
],
[
39,
50
],
[
51,
60
],
[
61,
63
],
[
64,
68
],
[
69,
72
],
[
73,
75
],
[
76,
... |
ai-train-60 | [
[
6,
6,
"programlang"
],
[
8,
8,
"product"
],
[
10,
11,
"programlang"
],
[
15,
15,
"product"
],
[
21,
21,
"programlang"
]
] | [
0,
1,
2,
3,
4
] | [
[
6,
6,
10,
11,
"type-of",
"",
false,
false
],
[
6,
6,
21,
21,
"named",
"",
false,
false
],
[
8,
8,
10,
11,
"part-of",
"",
false,
false
],
[
8,
8,
15,
15,
"named",
"",
false,
f... | [
0,
1,
2,
3
] | [
"Même",
"implémentation",
",",
"à",
"exécuter",
"en",
"Java",
"avec",
"JShell",
"(",
"Java",
"9",
"minimum",
")",
":",
"codejshell",
"scriptfile",
"/",
"codesyntaxhighlight",
"lang",
"=",
"java"
] | Même implémentation, à exécuter en Java avec JShell (Java 9 minimum) : codejshell scriptfile / codesyntaxhighlight lang = java | [
[
0,
4
],
[
5,
19
],
[
19,
20
],
[
21,
22
],
[
23,
31
],
[
32,
34
],
[
35,
39
],
[
40,
44
],
[
45,
51
],
[
52,
53
],
[
53,
57
],
[
58,
59
],
[
60,
67
],
[
6... |
ai-train-61 | [
[
0,
3,
"metrics"
],
[
8,
9,
"metrics"
]
] | [
0,
1
] | [
[
0,
3,
8,
9,
"origin",
"based_on",
false,
false
]
] | [
0
] | [
"La",
"métrique",
"du",
"NIST",
"est",
"basée",
"sur",
"la",
"métrique",
"BLEU",
",",
"mais",
"avec",
"quelques",
"modifications",
"."
] | La métrique du NIST est basée sur la métrique BLEU, mais avec quelques modifications. | [
[
0,
2
],
[
3,
11
],
[
12,
14
],
[
15,
19
],
[
20,
23
],
[
24,
29
],
[
30,
33
],
[
34,
36
],
[
37,
45
],
[
46,
50
],
[
50,
51
],
[
52,
56
],
[
57,
61
],
[
6... |
ai-train-62 | [
[
9,
9,
"country"
],
[
12,
14,
"university"
],
[
17,
19,
"university"
],
[
27,
28,
"product"
],
[
33,
34,
"algorithm"
]
] | [
0,
1,
2,
3,
4
] | [
[
12,
14,
9,
9,
"physical",
"",
false,
false
],
[
17,
19,
9,
9,
"physical",
"",
false,
false
],
[
27,
28,
12,
14,
"origin",
"",
false,
false
],
[
27,
28,
17,
19,
"origin",
"",
fals... | [
0,
1,
2,
3,
4
] | [
"À",
"la",
"fin",
"des",
"années",
"1980",
",",
"deux",
"universités",
"néerlandaises",
",",
"l'",
"Université",
"de",
"Groningue",
"et",
"l'",
"Université",
"de",
"Twente",
",",
"ont",
"lancé",
"conjointement",
"un",
"projet",
"appelé",
"Knowledge",
"Graphs",... | À la fin des années 1980, deux universités néerlandaises, l'Université de Groningue et l'Université de Twente, ont lancé conjointement un projet appelé Knowledge Graphs, qui sont des réseaux sémantiques, mais avec la contrainte supplémentaire que les arêtes sont limitées à un ensemble restreint de relations possibles, ... | [
[
0,
1
],
[
2,
4
],
[
5,
8
],
[
9,
12
],
[
13,
19
],
[
20,
24
],
[
24,
25
],
[
26,
30
],
[
31,
42
],
[
43,
56
],
[
56,
57
],
[
58,
60
],
[
60,
70
],
[
71,
... |
ai-train-63 | [
[
1,
2,
"product"
],
[
17,
19,
"product"
]
] | [
0,
1
] | [
[
1,
2,
17,
19,
"part-of",
"",
false,
false
]
] | [
0
] | [
"Les",
"correcteurs",
"grammaticaux",
"sont",
"le",
"plus",
"souvent",
"intégrés",
"à",
"un",
"programme",
"plus",
"vaste",
",",
"tel",
"qu'",
"un",
"traitement",
"de",
"texte",
",",
"mais",
"ils",
"sont",
"également",
"disponibles",
"sous",
"la",
"forme",
"... | Les correcteurs grammaticaux sont le plus souvent intégrés à un programme plus vaste, tel qu'un traitement de texte, mais ils sont également disponibles sous la forme d'une application autonome qui peut être activée à partir de programmes qui traitent du texte éditable. | [
[
0,
3
],
[
4,
15
],
[
16,
28
],
[
29,
33
],
[
34,
36
],
[
37,
41
],
[
42,
49
],
[
50,
58
],
[
59,
60
],
[
61,
63
],
[
64,
73
],
[
74,
78
],
[
79,
84
],
[
8... |
ai-train-64 | [
[
5,
11,
"organisation"
],
[
15,
20,
"conference"
],
[
24,
26,
"organisation"
],
[
33,
35,
"conference"
],
[
37,
39,
"conference"
],
[
41,
43,
"conference"
]
] | [
0,
1,
2,
3,
4,
5
] | [] | [] | [
"Il",
"est",
"membre",
"de",
"l'",
"American",
"Association",
"for",
"the",
"Advancement",
"of",
"Science",
",",
"de",
"l'",
"Association",
"for",
"the",
"Advancement",
"Artificial",
"Intelligence",
"et",
"de",
"la",
"Cognitive",
"Science",
"Society",
",",
"et... | Il est membre de l'American Association for the Advancement of Science, de l'Association for the Advancement Artificial Intelligence et de la Cognitive Science Society, et rédacteur en chef de J. Automated Reasoning, J. Learning Sciences et J. Applied Ontology. | [
[
0,
2
],
[
3,
6
],
[
7,
13
],
[
14,
16
],
[
17,
19
],
[
19,
27
],
[
28,
39
],
[
40,
43
],
[
44,
47
],
[
48,
59
],
[
60,
62
],
[
63,
70
],
[
70,
71
],
[
72,... |
ai-train-65 | [
[
0,
3,
"algorithm"
],
[
5,
5,
"algorithm"
],
[
11,
14,
"task"
],
[
25,
26,
"researcher"
],
[
29,
31,
"university"
],
[
34,
35,
"researcher"
],
[
37,
40,
"organisation"
],
[
42,
42,
"organisat... | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
[
0,
3,
11,
14,
"type-of",
"",
false,
false
],
[
0,
3,
25,
26,
"origin",
"",
false,
false
],
[
0,
3,
34,
35,
"origin",
"",
false,
false
],
[
5,
5,
0,
3,
"named",
"",
false,
fal... | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
"Le",
"codage",
"prédictif",
"linéaire",
"(",
"LPC",
")",
",",
"une",
"forme",
"de",
"codage",
"de",
"la",
"parole",
",",
"a",
"commencé",
"à",
"se",
"développer",
"avec",
"les",
"travaux",
"de",
"Fumitada",
"Itakura",
"de",
"l'",
"université",
"de",
"Na... | Le codage prédictif linéaire (LPC), une forme de codage de la parole, a commencé à se développer avec les travaux de Fumitada Itakura de l'université de Nagoya et de Shuzo Saito de Nippon Telegraph and Telephone (NTT) en 1966. | [
[
0,
2
],
[
3,
9
],
[
10,
19
],
[
20,
28
],
[
29,
30
],
[
30,
33
],
[
33,
34
],
[
34,
35
],
[
36,
39
],
[
40,
45
],
[
46,
48
],
[
49,
55
],
[
56,
58
],
[
59... |
ai-train-66 | [
[
56,
58,
"metrics"
]
] | [
0
] | [] | [] | [
"Si",
"le",
"signal",
"est",
"en",
"outre",
"ergodique",
",",
"tous",
"les",
"chemins",
"d'",
"échantillonnage",
"présentent",
"la",
"même",
"moyenne",
"temporelle",
"et",
"donc",
"mathR",
"_",
"x",
"^",
"{n",
"/",
"T",
"_",
"0}",
"(",
"\\",
"tau",
")"... | Si le signal est en outre ergodique, tous les chemins d'échantillonnage présentent la même moyenne temporelle et donc mathR _ x ^ {n / T _ 0} (\ tau) =\ widehat {R} _ x ^ {n / T _ 0} (\ tau) / math au sens de l'erreur quadratique moyenne. | [
[
0,
2
],
[
3,
5
],
[
6,
12
],
[
13,
16
],
[
17,
19
],
[
20,
25
],
[
26,
35
],
[
35,
36
],
[
37,
41
],
[
42,
45
],
[
46,
53
],
[
54,
56
],
[
56,
71
],
[
72,... |
ai-train-67 | [
[
0,
3,
"task"
],
[
6,
9,
"task"
],
[
22,
25,
"algorithm"
],
[
27,
27,
"algorithm"
],
[
32,
34,
"algorithm"
],
[
36,
36,
"algorithm"
],
[
41,
44,
"algorithm"
],
[
46,
46,
"algorithm"
],
[
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12
] | [
[
22,
25,
62,
64,
"related-to",
"",
false,
false
],
[
27,
27,
22,
25,
"named",
"",
false,
false
],
[
32,
34,
62,
64,
"related-to",
"",
false,
false
],
[
36,
36,
32,
34,
"named",
"",
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
"L'",
"extraction",
"de",
"caractéristiques",
"et",
"la",
"réduction",
"de",
"la",
"dimension",
"peuvent",
"être",
"combinées",
"en",
"une",
"seule",
"étape",
"à",
"l'",
"aide",
"de",
"l'",
"analyse",
"en",
"composantes",
"principales",
"(",
"ACP",
")",
",",... | L'extraction de caractéristiques et la réduction de la dimension peuvent être combinées en une seule étape à l'aide de l'analyse en composantes principales (ACP), de l'analyse discriminante linéaire (ADL), de l'analyse de corrélation canonique (ACC) ou des techniques de factorisation de la matrice non négative (NMF) co... | [
[
0,
2
],
[
2,
12
],
[
13,
15
],
[
16,
32
],
[
33,
35
],
[
36,
38
],
[
39,
48
],
[
49,
51
],
[
52,
54
],
[
55,
64
],
[
65,
72
],
[
73,
77
],
[
78,
87
],
[
8... |
ai-train-68 | [
[
4,
4,
"programlang"
],
[
6,
6,
"programlang"
],
[
8,
8,
"programlang"
],
[
10,
10,
"programlang"
],
[
16,
16,
"product"
]
] | [
0,
1,
2,
3,
4
] | [
[
16,
16,
4,
4,
"related-to",
"program_type_compatible_with",
false,
false
],
[
16,
16,
6,
6,
"related-to",
"program_type_compatible_with",
false,
false
],
[
16,
16,
8,
8,
"related-to",
"program_type_compatible_with",
... | [
0,
1,
2,
3
] | [
"Les",
"bibliothèques",
"écrites",
"en",
"Perl",
",",
"Java",
",",
"ActiveX",
"ou",
".NET",
"peuvent",
"être",
"appelées",
"directement",
"depuis",
"MATLAB",
","
] | Les bibliothèques écrites en Perl, Java, ActiveX ou .NET peuvent être appelées directement depuis MATLAB, | [
[
0,
3
],
[
4,
17
],
[
18,
25
],
[
26,
28
],
[
29,
33
],
[
33,
34
],
[
35,
39
],
[
39,
40
],
[
41,
48
],
[
49,
51
],
[
52,
56
],
[
57,
64
],
[
65,
69
],
[
7... |
ai-train-69 | [
[
4,
10,
"task"
],
[
13,
16,
"task"
],
[
36,
38,
"task"
]
] | [
0,
1,
2
] | [
[
4,
10,
13,
16,
"part-of",
"",
false,
false
]
] | [
0
] | [
"La",
"tâche",
"consistant",
"à",
"reconnaître",
"des",
"entités",
"nommées",
"dans",
"un",
"texte",
"est",
"appelée",
"reconnaissance",
"d'",
"entités",
"nommées",
",",
"tandis",
"que",
"la",
"tâche",
"consistant",
"à",
"déterminer",
"l'",
"identité",
"des",
... | La tâche consistant à reconnaître des entités nommées dans un texte est appelée reconnaissance d'entités nommées, tandis que la tâche consistant à déterminer l'identité des entités nommées mentionnées dans un texte est appelée liaison d'entités. | [
[
0,
2
],
[
3,
8
],
[
9,
19
],
[
20,
21
],
[
22,
33
],
[
34,
37
],
[
38,
45
],
[
46,
53
],
[
54,
58
],
[
59,
61
],
[
62,
67
],
[
68,
71
],
[
72,
79
],
[
80,... |
ai-train-70 | [
[
1,
2,
"algorithm"
],
[
33,
33,
"programlang"
],
[
35,
35,
"algorithm"
]
] | [
0,
1,
2
] | [
[
1,
2,
35,
35,
"part-of",
"",
true,
false
],
[
35,
35,
33,
33,
"part-of",
"",
true,
false
]
] | [
0,
1
] | [
"Les",
"fonctions",
"sigmoïdes",
"et",
"les",
"dérivés",
"utilisés",
"dans",
"le",
"paquetage",
"étaient",
"à",
"l'",
"origine",
"inclus",
"dans",
"le",
"paquetage",
",",
"à",
"partir",
"de",
"la",
"version",
"0.8.0",
",",
"ils",
"ont",
"été",
"publiés",
"... | Les fonctions sigmoïdes et les dérivés utilisés dans le paquetage étaient à l'origine inclus dans le paquetage, à partir de la version 0.8.0, ils ont été publiés dans un paquetage R séparé sigmoïde, dans le but de permettre une utilisation plus générale. | [
[
0,
3
],
[
4,
13
],
[
14,
23
],
[
24,
26
],
[
27,
30
],
[
31,
38
],
[
39,
47
],
[
48,
52
],
[
53,
55
],
[
56,
65
],
[
66,
73
],
[
74,
75
],
[
76,
78
],
[
7... |
ai-train-71 | [
[
1,
1,
"programlang"
],
[
8,
12,
"organisation"
],
[
14,
14,
"organisation"
],
[
22,
22,
"location"
],
[
24,
24,
"location"
],
[
27,
28,
"researcher"
],
[
30,
31,
"researcher"
],
[
33,
34,
"r... | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
[
1,
1,
27,
28,
"artifact",
"",
true,
false
],
[
1,
1,
30,
31,
"artifact",
"",
true,
false
],
[
1,
1,
33,
34,
"artifact",
"",
true,
false
],
[
14,
14,
8,
12,
"named",
"",
false,
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
"Le",
"logo",
"a",
"été",
"créé",
"en",
"1967",
"chez",
"Bolt",
",",
"Beranek",
"et",
"Newman",
"(",
"BBN",
")",
",",
"un",
"cabinet",
"de",
"recherche",
"de",
"Cambridge",
",",
"Massachusetts",
",",
"par",
"Wally",
"Feurzeig",
",",
"Cynthia",
"Solomon",... | Le logo a été créé en 1967 chez Bolt, Beranek et Newman (BBN), un cabinet de recherche de Cambridge, Massachusetts, par Wally Feurzeig, Cynthia Solomon et Seymour Papert. | [
[
0,
2
],
[
3,
7
],
[
8,
9
],
[
10,
13
],
[
14,
18
],
[
19,
21
],
[
22,
26
],
[
27,
31
],
[
32,
36
],
[
36,
37
],
[
38,
45
],
[
46,
48
],
[
49,
55
],
[
56,
... |
ai-train-72 | [
[
0,
1,
"misc"
],
[
12,
14,
"field"
],
[
25,
26,
"field"
],
[
30,
32,
"algorithm"
],
[
35,
36,
"algorithm"
]
] | [
0,
1,
2,
3,
4
] | [
[
0,
1,
12,
14,
"part-of",
"",
false,
false
],
[
0,
1,
25,
26,
"compare",
"",
false,
false
],
[
30,
32,
25,
26,
"part-of",
"",
false,
false
],
[
35,
36,
25,
26,
"part-of",
"",
fals... | [
0,
1,
2,
3
] | [
"La",
"neuroévolution",
"est",
"couramment",
"utilisée",
"dans",
"le",
"cadre",
"du",
"paradigme",
"de",
"l'",
"apprentissage",
"par",
"renforcement",
",",
"et",
"elle",
"peut",
"être",
"comparée",
"aux",
"techniques",
"conventionnelles",
"d'",
"apprentissage",
"p... | La neuroévolution est couramment utilisée dans le cadre du paradigme de l'apprentissage par renforcement, et elle peut être comparée aux techniques conventionnelles d'apprentissage profond qui utilisent la descente de gradient sur un réseau neuronal à topologie fixe. | [
[
0,
2
],
[
3,
17
],
[
18,
21
],
[
22,
32
],
[
33,
41
],
[
42,
46
],
[
47,
49
],
[
50,
55
],
[
56,
58
],
[
59,
68
],
[
69,
71
],
[
72,
74
],
[
74,
87
],
[
8... |
ai-train-73 | [
[
4,
5,
"algorithm"
],
[
57,
59,
"metrics"
],
[
61,
61,
"metrics"
]
] | [
0,
1,
2
] | [
[
61,
61,
57,
59,
"named",
"",
false,
false
]
] | [
0
] | [
"Si",
"nous",
"utilisons",
"les",
"moindres",
"carrés",
"pour",
"ajuster",
"une",
"fonction",
"sous",
"la",
"forme",
"d'",
"un",
"hyperplan",
"ŷ",
"=",
"a",
"+",
"β",
"supT",
"/",
"sup",
"x",
"aux",
"données",
"(",
"x",
"sub",
"i",
"/",
"sub",
",",
... | Si nous utilisons les moindres carrés pour ajuster une fonction sous la forme d'un hyperplan ŷ = a + β supT / sup x aux données (x sub i / sub, y sub i / sub) sub 1 ≤ i ≤n / sub, nous pourrions alors évaluer l'ajustement en utilisant l'erreur quadratique moyenne (EQM). | [
[
0,
2
],
[
3,
7
],
[
8,
17
],
[
18,
21
],
[
22,
30
],
[
31,
37
],
[
38,
42
],
[
43,
50
],
[
51,
54
],
[
55,
63
],
[
64,
68
],
[
69,
71
],
[
72,
77
],
[
78,... |
ai-train-74 | [
[
7,
7,
"country"
],
[
9,
10,
"country"
],
[
13,
13,
"country"
],
[
16,
16,
"country"
],
[
19,
19,
"country"
],
[
22,
22,
"country"
],
[
25,
25,
"country"
],
[
28,
28,
"country"
],
[
3... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20
] | [] | [] | [
"L'",
"entreprise",
"dispose",
"de",
"sites",
"internationaux",
"en",
"Australie",
",",
"au",
"Brésil",
",",
"au",
"Canada",
",",
"en",
"Chine",
",",
"en",
"Allemagne",
",",
"en",
"Inde",
",",
"en",
"Italie",
",",
"au",
"Japon",
",",
"en",
"Corée",
","... | L'entreprise dispose de sites internationaux en Australie, au Brésil, au Canada, en Chine, en Allemagne, en Inde, en Italie, au Japon, en Corée, en Lituanie, en Pologne, en Malaisie, aux Philippines, en Russie, à Singapour, en Afrique du Sud, en Espagne, à Taïwan, en Thaïlande, en Turquie et au Royaume-Uni. | [
[
0,
2
],
[
2,
12
],
[
13,
20
],
[
21,
23
],
[
24,
29
],
[
30,
44
],
[
45,
47
],
[
48,
57
],
[
57,
58
],
[
59,
61
],
[
62,
68
],
[
68,
69
],
[
70,
72
],
[
7... |
ai-train-75 | [
[
3,
3,
"misc"
],
[
6,
9,
"field"
],
[
15,
15,
"organisation"
],
[
19,
26,
"university"
],
[
32,
34,
"organisation"
],
[
38,
43,
"university"
],
[
50,
51,
"university"
],
[
55,
56,
"university... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
[
3,
3,
6,
9,
"topic",
"",
false,
false
],
[
3,
3,
15,
15,
"origin",
"",
false,
false
],
[
3,
3,
19,
26,
"origin",
"",
false,
false
]
] | [
0,
1,
2
] | [
"Titulaire",
"d'",
"un",
"D.Sc",
".",
"en",
"génie",
"électrique",
"et",
"informatique",
"(",
"2000",
")",
"de",
"l'",
"Inria",
"et",
"de",
"l'",
"Université",
"de",
"Nice",
"Sophia",
"Antipolis",
",",
"il",
"a",
"occupé",
"des",
"postes",
"permanents",
... | Titulaire d'un D.Sc. en génie électrique et informatique (2000) de l'Inria et de l'Université de Nice Sophia Antipolis, il a occupé des postes permanents chez Siemens Corporate Technology, à l'École des ponts ParisTech ainsi que des postes de visiteur à l'Université Rutgers, à l'Université Yale et à l'Université de Hou... | [
[
0,
9
],
[
10,
12
],
[
12,
14
],
[
15,
19
],
[
19,
20
],
[
21,
23
],
[
24,
29
],
[
30,
40
],
[
41,
43
],
[
44,
56
],
[
57,
58
],
[
58,
62
],
[
62,
63
],
[
... |
ai-train-76 | [
[
11,
12,
"researcher"
],
[
14,
14,
"researcher"
],
[
19,
20,
"product"
],
[
22,
22,
"country"
],
[
25,
25,
"product"
]
] | [
0,
1,
2,
3,
4
] | [
[
14,
14,
11,
12,
"role",
"licensing_patent_to",
false,
false
],
[
14,
14,
22,
22,
"physical",
"",
false,
false
],
[
25,
25,
14,
14,
"artifact",
"",
false,
false
],
[
25,
25,
19,
20,
"type... | [
0,
1,
2,
3
] | [
"Détenteur",
"d'",
"une",
"licence",
"du",
"brevet",
"original",
"accordé",
"à",
"l'",
"inventeur",
"George",
"Devol",
",",
"Engelberger",
"a",
"développé",
"le",
"premier",
"robot",
"industriel",
"aux",
"États-Unis",
",",
"l'",
"Unimate",
",",
"dans",
"les",
... | Détenteur d'une licence du brevet original accordé à l'inventeur George Devol, Engelberger a développé le premier robot industriel aux États-Unis, l'Unimate, dans les années 1950. | [
[
0,
9
],
[
10,
12
],
[
12,
15
],
[
16,
23
],
[
24,
26
],
[
27,
33
],
[
34,
42
],
[
43,
50
],
[
51,
52
],
[
53,
55
],
[
55,
64
],
[
65,
71
],
[
72,
77
],
[
... |
ai-train-77 | [
[
4,
5,
"task"
],
[
11,
12,
"task"
]
] | [
0,
1
] | [] | [] | [
"L'",
"entrée",
"est",
"appelée",
"reconnaissance",
"vocale",
"et",
"la",
"sortie",
"est",
"appelée",
"synthèse",
"vocale",
"."
] | L'entrée est appelée reconnaissance vocale et la sortie est appelée synthèse vocale. | [
[
0,
2
],
[
2,
8
],
[
9,
12
],
[
13,
20
],
[
21,
35
],
[
36,
42
],
[
43,
45
],
[
46,
48
],
[
49,
55
],
[
56,
59
],
[
60,
67
],
[
68,
76
],
[
77,
83
],
[
83,... |
ai-train-78 | [
[
5,
5,
"programlang"
],
[
8,
8,
"programlang"
],
[
12,
12,
"programlang"
],
[
19,
19,
"programlang"
],
[
31,
31,
"programlang"
]
] | [
0,
1,
2,
3,
4
] | [
[
5,
5,
12,
12,
"named",
"",
false,
false
],
[
8,
8,
5,
5,
"origin",
"descendant_of",
false,
false
],
[
8,
8,
19,
19,
"general-affiliation",
"",
false,
false
],
[
8,
8,
31,
31,
"named",
... | [
0,
1,
2,
3
] | [
"Parmi",
"les",
"descendants",
"du",
"langage",
"CLIPS",
",",
"citons",
"Jess",
"(",
"partie",
"de",
"CLIPS",
"basée",
"sur",
"les",
"règles",
"réécrite",
"en",
"Java",
",",
"elle",
"a",
"ensuite",
"évolué",
"dans",
"une",
"autre",
"direction",
")",
",",
... | Parmi les descendants du langage CLIPS, citons Jess (partie de CLIPS basée sur les règles réécrite en Java, elle a ensuite évolué dans une autre direction), JESS a été inspiré à l'origine de | [
[
0,
5
],
[
6,
9
],
[
10,
21
],
[
22,
24
],
[
25,
32
],
[
33,
38
],
[
38,
39
],
[
40,
46
],
[
47,
51
],
[
52,
53
],
[
53,
59
],
[
60,
62
],
[
63,
68
],
[
69... |
ai-train-79 | [
[
6,
6,
"product"
],
[
13,
16,
"product"
],
[
19,
20,
"organisation"
],
[
24,
25,
"product"
],
[
47,
49,
"product"
],
[
52,
55,
"product"
],
[
79,
81,
"misc"
]
] | [
0,
1,
2,
3,
4,
5,
6
] | [
[
13,
16,
6,
6,
"type-of",
"",
false,
false
],
[
19,
20,
13,
16,
"usage",
"",
false,
false
],
[
24,
25,
19,
20,
"artifact",
"",
false,
false
],
[
47,
49,
19,
20,
"origin",
"",
true... | [
0,
1,
2,
3,
4,
5,
6
] | [
"Elle",
"a",
"également",
"créé",
"des",
"applications",
"AGV",
"intelligentes",
"flexibles",
",",
"en",
"concevant",
"le",
"système",
"de",
"commande",
"Motivity",
"utilisé",
"par",
"RMT",
"Robotics",
"pour",
"développer",
"son",
"ADAM",
"iAGV",
"(",
"Self",
... | Elle a également créé des applications AGV intelligentes flexibles, en concevant le système de commande Motivity utilisé par RMT Robotics pour développer son ADAM iAGV (Self-Guided Vehicle), utilisé pour des opérations complexes de prélèvement et de placement, en conjonction avec des systèmes de portique et des bras de... | [
[
0,
4
],
[
5,
6
],
[
7,
16
],
[
17,
21
],
[
22,
25
],
[
26,
38
],
[
39,
42
],
[
43,
56
],
[
57,
66
],
[
66,
67
],
[
68,
70
],
[
71,
80
],
[
81,
83
],
[
84,... |
ai-train-80 | [
[
7,
9,
"metrics"
]
] | [
0
] | [] | [] | [
"Les",
"paramètres",
"β",
"sont",
"généralement",
"estimés",
"par",
"maximum",
"de",
"vraisemblance",
"."
] | Les paramètres β sont généralement estimés par maximum de vraisemblance. | [
[
0,
3
],
[
4,
14
],
[
15,
16
],
[
17,
21
],
[
22,
34
],
[
35,
42
],
[
43,
46
],
[
47,
54
],
[
55,
57
],
[
58,
71
],
[
71,
72
]
] |
ai-train-81 | [
[
3,
5,
"task"
],
[
9,
9,
"metrics"
],
[
12,
12,
"metrics"
],
[
15,
15,
"metrics"
]
] | [
0,
1,
2,
3
] | [
[
9,
9,
3,
5,
"part-of",
"",
false,
false
],
[
12,
12,
3,
5,
"part-of",
"",
false,
false
],
[
15,
15,
3,
5,
"part-of",
"",
false,
false
]
] | [
0,
1,
2
] | [
"Les",
"métriques",
"de",
"recherche",
"d'",
"information",
"telles",
"que",
"la",
"précision",
"et",
"le",
"rappel",
"ou",
"le",
"DCG",
"sont",
"utiles",
"pour",
"évaluer",
"la",
"qualité",
"d'",
"une",
"méthode",
"de",
"recommandation",
"."
] | Les métriques de recherche d'information telles que la précision et le rappel ou le DCG sont utiles pour évaluer la qualité d'une méthode de recommandation. | [
[
0,
3
],
[
4,
13
],
[
14,
16
],
[
17,
26
],
[
27,
29
],
[
29,
40
],
[
41,
47
],
[
48,
51
],
[
52,
54
],
[
55,
64
],
[
65,
67
],
[
68,
70
],
[
71,
77
],
[
7... |
ai-train-82 | [
[
7,
8,
"product"
]
] | [
0
] | [] | [] | [
"Une",
"usine",
"typique",
"contient",
"des",
"centaines",
"de",
"robots",
"industriels",
"travaillant",
"sur",
"des",
"lignes",
"de",
"production",
"entièrement",
"automatisées",
",",
"avec",
"un",
"robot",
"pour",
"dix",
"travailleurs",
"humains",
"."
] | Une usine typique contient des centaines de robots industriels travaillant sur des lignes de production entièrement automatisées, avec un robot pour dix travailleurs humains. | [
[
0,
3
],
[
4,
9
],
[
10,
17
],
[
18,
26
],
[
27,
30
],
[
31,
40
],
[
41,
43
],
[
44,
50
],
[
51,
62
],
[
63,
74
],
[
75,
78
],
[
79,
82
],
[
83,
89
],
[
90... |
ai-train-83 | [
[
8,
10,
"product"
],
[
16,
18,
"field"
],
[
22,
24,
"task"
],
[
27,
29,
"task"
],
[
32,
34,
"task"
],
[
37,
39,
"task"
],
[
42,
44,
"task"
],
[
47,
49,
"task"
]
] | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
[
16,
18,
8,
10,
"usage",
"",
false,
true
],
[
22,
24,
16,
18,
"part-of",
"",
false,
false
],
[
27,
29,
16,
18,
"part-of",
"",
false,
false
],
[
32,
34,
16,
18,
"part-of",
"",
fals... | [
0,
1,
2,
3,
4,
5,
6
] | [
"Au",
"cours",
"de",
"la",
"dernière",
"décennie",
",",
"les",
"PCNN",
"ont",
"été",
"utilisés",
"dans",
"diverses",
"applications",
"de",
"traitement",
"d'",
"images",
",",
"notamment",
"la",
"segmentation",
"d'",
"images",
",",
"la",
"génération",
"de",
"c... | Au cours de la dernière décennie, les PCNN ont été utilisés dans diverses applications de traitement d'images, notamment la segmentation d'images, la génération de caractéristiques, l'extraction de visages, la détection de mouvements, la croissance de régions et la réduction du bruit. | [
[
0,
2
],
[
3,
8
],
[
9,
11
],
[
12,
14
],
[
15,
23
],
[
24,
32
],
[
32,
33
],
[
34,
37
],
[
38,
42
],
[
43,
46
],
[
47,
50
],
[
51,
59
],
[
60,
64
],
[
65,... |
ai-train-84 | [
[
0,
0,
"researcher"
],
[
19,
21,
"field"
],
[
26,
30,
"misc"
],
[
32,
41,
"conference"
],
[
43,
43,
"conference"
],
[
48,
51,
"misc"
],
[
53,
59,
"conference"
],
[
60,
61,
"conference"
],
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9
] | [
[
0,
0,
19,
21,
"related-to",
"contributes_to",
false,
false
],
[
0,
0,
26,
30,
"win-defeat",
"",
false,
false
],
[
0,
0,
48,
51,
"win-defeat",
"",
false,
false
],
[
26,
30,
32,
41,
"tempo... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
"Xu",
"a",
"publié",
"plus",
"de",
"50",
"articles",
"dans",
"des",
"conférences",
"internationales",
"et",
"des",
"revues",
"dans",
"le",
"domaine",
"de",
"la",
"vision",
"par",
"ordinateur",
"et",
"a",
"remporté",
"le",
"prix",
"du",
"meilleur",
"article",... | Xu a publié plus de 50 articles dans des conférences internationales et des revues dans le domaine de la vision par ordinateur et a remporté le prix du meilleur article à la conférence internationale sur le rendu et l'animation non photoréalistes (NPAR) 2012 et le prix du meilleur réviseur aux conférences international... | [
[
0,
2
],
[
3,
4
],
[
5,
11
],
[
12,
16
],
[
17,
19
],
[
20,
22
],
[
23,
31
],
[
32,
36
],
[
37,
40
],
[
41,
52
],
[
53,
68
],
[
69,
71
],
[
72,
75
],
[
76,... |
ai-train-85 | [
[
0,
0,
"programlang"
],
[
2,
2,
"field"
],
[
5,
6,
"field"
],
[
9,
10,
"misc"
],
[
18,
19,
"researcher"
],
[
14,
16,
"misc"
]
] | [
0,
1,
2,
3,
4,
5
] | [
[
0,
0,
2,
2,
"part-of",
"",
false,
false
],
[
0,
0,
5,
6,
"part-of",
"",
false,
false
],
[
0,
0,
9,
10,
"type-of",
"",
false,
false
],
[
14,
16,
0,
0,
"usage",
"",
false,
fals... | [
0,
1,
2,
3,
4
] | [
"CycL",
"en",
"informatique",
"et",
"en",
"intelligence",
"artificielle",
"est",
"un",
"langage",
"ontologique",
"utilisé",
"par",
"le",
"projet",
"Cyc",
"artificial",
"de",
"Doug",
"Lenat",
"."
] | CycL en informatique et en intelligence artificielle est un langage ontologique utilisé par le projet Cyc artificial de Doug Lenat. | [
[
0,
4
],
[
5,
7
],
[
8,
20
],
[
21,
23
],
[
24,
26
],
[
27,
39
],
[
40,
52
],
[
53,
56
],
[
57,
59
],
[
60,
67
],
[
68,
79
],
[
80,
87
],
[
88,
91
],
[
92,... |
ai-train-86 | [
[
3,
5,
"task"
],
[
8,
10,
"metrics"
],
[
14,
18,
"metrics"
],
[
20,
24,
"metrics"
],
[
32,
33,
"misc"
]
] | [
0,
1,
2,
3,
4
] | [
[
8,
10,
3,
5,
"part-of",
"",
false,
false
],
[
14,
18,
8,
10,
"named",
"",
false,
false
],
[
20,
24,
8,
10,
"named",
"",
false,
false
],
[
32,
33,
8,
10,
"named",
"",
false,
f... | [
0,
1,
2,
3
] | [
"Toujours",
"dans",
"l'",
"analyse",
"de",
"régression",
",",
"l'",
"erreur",
"quadratique",
"moyenne",
",",
"souvent",
"appelée",
"erreur",
"quadratique",
"moyenne",
"de",
"prédiction",
"ou",
"erreur",
"quadratique",
"moyenne",
"hors",
"échantillon",
",",
"peut",... | Toujours dans l'analyse de régression, l'erreur quadratique moyenne, souvent appelée erreur quadratique moyenne de prédiction ou erreur quadratique moyenne hors échantillon, peut désigner la valeur moyenne des écarts quadratiques des prédictions par rapport aux valeurs VRAIES, sur un espace de test hors échantillon, gé... | [
[
0,
8
],
[
9,
13
],
[
14,
16
],
[
16,
23
],
[
24,
26
],
[
27,
37
],
[
37,
38
],
[
39,
41
],
[
41,
47
],
[
48,
59
],
[
60,
67
],
[
67,
68
],
[
69,
76
],
[
7... |
ai-train-87 | [
[
11,
11,
"algorithm"
],
[
13,
13,
"algorithm"
],
[
20,
21,
"algorithm"
],
[
37,
39,
"metrics"
]
] | [
0,
1,
2,
3
] | [
[
11,
11,
13,
13,
"compare",
"",
false,
false
],
[
11,
11,
20,
21,
"named",
"",
false,
false
]
] | [
0,
1
] | [
"En",
"ce",
"qui",
"concerne",
"les",
"résultats",
",",
"les",
"descripteurs",
"de",
"blocs",
"C-HOG",
"et",
"R-HOG",
"ont",
"des",
"performances",
"comparables",
",",
"les",
"descripteurs",
"C-HOG",
"conservant",
"un",
"léger",
"avantage",
"dans",
"le",
"taux... | En ce qui concerne les résultats, les descripteurs de blocs C-HOG et R-HOG ont des performances comparables, les descripteurs C-HOG conservant un léger avantage dans le taux d'échec de la détection pour des taux de faux positifs fixes sur les deux ensembles de données. | [
[
0,
2
],
[
3,
5
],
[
6,
9
],
[
10,
18
],
[
19,
22
],
[
23,
32
],
[
32,
33
],
[
34,
37
],
[
38,
50
],
[
51,
53
],
[
54,
59
],
[
60,
65
],
[
66,
68
],
[
69,
... |
ai-train-88 | [
[
9,
12,
"algorithm"
],
[
15,
16,
"misc"
],
[
19,
21,
"algorithm"
],
[
24,
25,
"algorithm"
],
[
28,
30,
"algorithm"
],
[
33,
36,
"algorithm"
],
[
39,
42,
"algorithm"
],
[
45,
46,
"misc"
],
... | [
0,
1,
2,
3,
4,
5,
6,
7,
8
] | [
[
9,
12,
15,
16,
"usage",
"",
false,
false
],
[
19,
21,
45,
46,
"usage",
"",
false,
false
],
[
24,
25,
45,
46,
"usage",
"",
false,
false
],
[
28,
30,
45,
46,
"usage",
"",
false,
... | [
0,
1,
2,
3,
4,
5
] | [
"Les",
"algorithmes",
"de",
"reconnaissance",
"les",
"plus",
"répandus",
"sont",
"l'",
"analyse",
"en",
"composantes",
"principales",
"utilisant",
"les",
"faces",
"propres",
",",
"l'",
"analyse",
"discriminante",
"linéaire",
",",
"l'",
"appariement",
"élastique",
... | Les algorithmes de reconnaissance les plus répandus sont l'analyse en composantes principales utilisant les faces propres, l'analyse discriminante linéaire, l'appariement élastique utilisant l'algorithme de Fisherface, le modèle de Markov caché, l'apprentissage du sous-espace multilinéaire utilisant la représentation t... | [
[
0,
3
],
[
4,
15
],
[
16,
18
],
[
19,
33
],
[
34,
37
],
[
38,
42
],
[
43,
51
],
[
52,
56
],
[
57,
59
],
[
59,
66
],
[
67,
69
],
[
70,
81
],
[
82,
93
],
[
9... |
ai-train-89 | [
[
4,
12,
"misc"
],
[
27,
29,
"location"
],
[
48,
50,
"location"
],
[
67,
67,
"organisation"
]
] | [
0,
1,
2,
3
] | [
[
27,
29,
4,
12,
"temporal",
"",
false,
false
],
[
48,
50,
4,
12,
"temporal",
"",
false,
false
]
] | [
0,
1
] | [
"À",
"partir",
"de",
"l'",
"édition",
"2019",
"du",
"Festival",
"international",
"du",
"film",
"de",
"Toronto",
",",
"il",
"se",
"peut",
"désormais",
"que",
"les",
"films",
"ne",
"puissent",
"plus",
"être",
"projetés",
"au",
"Scotiabank",
"Theatre",
"Toronto... | À partir de l'édition 2019 du Festival international du film de Toronto, il se peut désormais que les films ne puissent plus être projetés au Scotiabank Theatre Toronto - l'un des principaux sites du festival - et qu'ils soient projetés ailleurs (notamment au TIFF Bell Lightbox et dans d'autres cinémas locaux) s'ils so... | [
[
0,
1
],
[
2,
8
],
[
9,
11
],
[
12,
14
],
[
14,
21
],
[
22,
26
],
[
27,
29
],
[
30,
38
],
[
39,
52
],
[
53,
55
],
[
56,
60
],
[
61,
63
],
[
64,
71
],
[
71,... |
ai-train-90 | [
[
0,
0,
"organisation"
],
[
6,
7,
"researcher"
],
[
3,
4,
"organisation"
],
[
16,
16,
"researcher"
],
[
27,
32,
"product"
],
[
49,
49,
"researcher"
],
[
44,
47,
"programlang"
]
] | [
0,
1,
2,
3,
4,
5,
6
] | [
[
0,
0,
3,
4,
"related-to",
"purchases",
false,
false
],
[
6,
7,
16,
16,
"named",
"same",
false,
false
],
[
6,
7,
49,
49,
"named",
"same",
false,
false
],
[
3,
4,
6,
7,
"origin",
"foun... | [
0,
1,
2,
3,
4,
5
] | [
"Unimation",
"a",
"acheté",
"Vicarm",
"Inc.",
"de",
"Victor",
"Scheinman",
"en",
"1977",
",",
"et",
"avec",
"l'",
"aide",
"de",
"Scheinman",
",",
"la",
"société",
"a",
"créé",
"et",
"commencé",
"à",
"produire",
"la",
"Machine",
"Universelle",
"Programmable",... | Unimation a acheté Vicarm Inc. de Victor Scheinman en 1977, et avec l'aide de Scheinman, la société a créé et commencé à produire la Machine Universelle Programmable pour l'Assemblage, un nouveau modèle de bras robotique, en utilisant le langage de programmation VAL de Scheinman. | [
[
0,
9
],
[
10,
11
],
[
12,
18
],
[
19,
25
],
[
26,
30
],
[
31,
33
],
[
34,
40
],
[
41,
50
],
[
51,
53
],
[
54,
58
],
[
58,
59
],
[
60,
62
],
[
63,
67
],
[
... |
ai-train-91 | [
[
0,
0,
"product"
],
[
4,
4,
"programlang"
],
[
9,
10,
"algorithm"
],
[
13,
18,
"product"
]
] | [
0,
1,
2,
3
] | [
[
0,
0,
4,
4,
"general-affiliation",
"",
false,
false
],
[
0,
0,
9,
10,
"origin",
"implementation_of",
false,
false
],
[
0,
0,
13,
18,
"part-of",
"",
false,
false
]
] | [
0,
1,
2
] | [
"J48",
"est",
"une",
"implémentation",
"Java",
"open",
"source",
"de",
"l'",
"algorithme",
"C4.5",
"dans",
"l'",
"outil",
"d'",
"exploration",
"de",
"données",
"Weka",
"."
] | J48 est une implémentation Java open source de l'algorithme C4.5 dans l'outil d'exploration de données Weka. | [
[
0,
3
],
[
4,
7
],
[
8,
11
],
[
12,
26
],
[
27,
31
],
[
32,
36
],
[
37,
43
],
[
44,
46
],
[
47,
49
],
[
49,
59
],
[
60,
64
],
[
65,
69
],
[
70,
72
],
[
72,... |
ai-train-92 | [
[
2,
3,
"metrics"
],
[
13,
14,
"product"
],
[
21,
28,
"misc"
]
] | [
0,
1,
2
] | [
[
2,
3,
13,
14,
"win-defeat",
"",
false,
false
]
] | [
0
] | [
"L'",
"article",
"SSIM",
"de",
"2004",
"a",
"été",
"cité",
"plus",
"de",
"20 000",
"fois",
"selon",
"Google",
"Scholar",
".",
"Il",
"a",
"également",
"reçu",
"le",
"prix",
"IEEE",
"Signal",
"Processing",
"Society",
"Sustained",
"Impact",
"Award",
"pour",
"... | L'article SSIM de 2004 a été cité plus de 20 000 fois selon Google Scholar. Il a également reçu le prix IEEE Signal Processing Society Sustained Impact Award pour 2016, qui indique qu'un article a un impact inhabituellement élevé pendant au moins 10 ans après sa publication. | [
[
0,
2
],
[
2,
9
],
[
10,
14
],
[
15,
17
],
[
18,
22
],
[
23,
24
],
[
25,
28
],
[
29,
33
],
[
34,
38
],
[
39,
41
],
[
42,
48
],
[
49,
53
],
[
54,
59
],
[
60... |
ai-train-93 | [
[
1,
2,
"task"
],
[
33,
34,
"product"
],
[
44,
46,
"product"
],
[
50,
50,
"organisation"
],
[
51,
51,
"product"
],
[
56,
56,
"organisation"
]
] | [
0,
1,
2,
3,
4,
5
] | [
[
1,
2,
50,
50,
"artifact",
"",
false,
false
],
[
33,
34,
1,
2,
"related-to",
"performs",
false,
false
],
[
33,
34,
44,
46,
"part-of",
"",
false,
false
],
[
50,
50,
56,
56,
"part-of",
... | [
0,
1,
2,
3
] | [
"La",
"synthèse",
"vocale",
"est",
"sur",
"le",
"point",
"d'",
"être",
"totalement",
"indiscernable",
"de",
"la",
"voix",
"d'",
"un",
"véritable",
"être",
"humain",
"avec",
"l'",
"introduction",
"en",
"2016",
"du",
"logiciel",
"d'",
"édition",
"et",
"de",
... | La synthèse vocale est sur le point d'être totalement indiscernable de la voix d'un véritable être humain avec l'introduction en 2016 du logiciel d'édition et de génération de voix Adobe Voco, un prototype destiné à faire partie de la suite Adobe Creative, et de DeepMind WaveNet, un prototype de Google. | [
[
0,
2
],
[
3,
11
],
[
12,
18
],
[
19,
22
],
[
23,
26
],
[
27,
29
],
[
30,
35
],
[
36,
38
],
[
38,
42
],
[
43,
53
],
[
54,
67
],
[
68,
70
],
[
71,
73
],
[
7... |
ai-train-94 | [
[
0,
1,
"researcher"
],
[
6,
10,
"organisation"
],
[
15,
21,
"organisation"
],
[
27,
27,
"conference"
],
[
32,
36,
"organisation"
]
] | [
0,
1,
2,
3,
4
] | [
[
0,
1,
6,
10,
"role",
"",
false,
false
],
[
0,
1,
15,
21,
"role",
"",
false,
false
],
[
0,
1,
27,
27,
"role",
"",
false,
false
],
[
0,
1,
32,
36,
"role",
"",
false,
false
]
... | [
0,
1,
2,
3
] | [
"M.",
"Poggio",
"est",
"membre",
"honoraire",
"du",
"programme",
"de",
"recherche",
"en",
"neurosciences",
",",
"membre",
"de",
"l'",
"Académie",
"américaine",
"des",
"arts",
"et",
"des",
"sciences",
",",
"membre",
"fondateur",
"de",
"l'",
"AAAI",
"et",
"mem... | M. Poggio est membre honoraire du programme de recherche en neurosciences, membre de l'Académie américaine des arts et des sciences, membre fondateur de l'AAAI et membre fondateur du McGovern Institute for Brain Research. | [
[
0,
2
],
[
3,
9
],
[
10,
13
],
[
14,
20
],
[
21,
30
],
[
31,
33
],
[
34,
43
],
[
44,
46
],
[
47,
56
],
[
57,
59
],
[
60,
73
],
[
73,
74
],
[
75,
81
],
[
82... |
ai-train-95 | [
[
11,
16,
"task"
],
[
15,
15,
"task"
],
[
22,
23,
"task"
],
[
31,
31,
"misc"
],
[
30,
32,
"misc"
]
] | [
0,
1,
2,
3,
4
] | [
[
11,
16,
22,
23,
"cause-effect",
"",
false,
false
],
[
15,
15,
22,
23,
"cause-effect",
"",
false,
false
],
[
30,
32,
22,
23,
"topic",
"",
false,
false
],
[
30,
32,
31,
31,
"general-affili... | [
0,
1,
2,
3
] | [
"Dans",
"les",
"années",
"90",
",",
"encouragées",
"par",
"les",
"succès",
"de",
"la",
"reconnaissance",
"et",
"de",
"la",
"synthèse",
"vocales",
",",
"les",
"recherches",
"sur",
"la",
"traduction",
"vocale",
"ont",
"débuté",
"avec",
"le",
"développement",
"... | Dans les années 90, encouragées par les succès de la reconnaissance et de la synthèse vocales, les recherches sur la traduction vocale ont débuté avec le développement du projet allemand Verbmobil. | [
[
0,
4
],
[
5,
8
],
[
9,
15
],
[
16,
18
],
[
18,
19
],
[
20,
31
],
[
32,
35
],
[
36,
39
],
[
40,
46
],
[
47,
49
],
[
50,
52
],
[
53,
67
],
[
68,
70
],
[
71,... |
ai-train-96 | [
[
3,
4,
"researcher"
],
[
8,
9,
"researcher"
],
[
11,
13,
"researcher"
],
[
16,
18,
"algorithm"
],
[
22,
24,
"algorithm"
],
[
29,
29,
"algorithm"
]
] | [
0,
1,
2,
3,
4,
5
] | [
[
3,
4,
8,
9,
"role",
"",
false,
false
],
[
16,
18,
3,
4,
"origin",
"",
false,
false
],
[
16,
18,
8,
9,
"origin",
"",
false,
false
],
[
16,
18,
11,
13,
"origin",
"",
false,
fal... | [
0,
1,
2,
3,
4,
5
] | [
"En",
"1999",
",",
"Felix",
"Gers",
"et",
"son",
"conseiller",
"Jürgen",
"Schmidhuber",
"et",
"Fred",
"Cummins",
"ont",
"introduit",
"la",
"porte",
"d'",
"oubli",
"(",
"également",
"appelée",
"porte",
"de",
"maintien",
")",
"dans",
"l'",
"architecture",
"LST... | En 1999, Felix Gers et son conseiller Jürgen Schmidhuber et Fred Cummins ont introduit la porte d'oubli (également appelée porte de maintien) dans l'architecture LSTM, | [
[
0,
2
],
[
3,
7
],
[
7,
8
],
[
9,
14
],
[
15,
19
],
[
20,
22
],
[
23,
26
],
[
27,
37
],
[
38,
44
],
[
45,
56
],
[
57,
59
],
[
60,
64
],
[
65,
72
],
[
73,
... |
ai-train-97 | [
[
2,
5,
"field"
],
[
8,
11,
"field"
],
[
14,
16,
"algorithm"
]
] | [
0,
1,
2
] | [
[
14,
16,
2,
5,
"part-of",
"",
false,
false
],
[
14,
16,
8,
11,
"part-of",
"",
false,
false
]
] | [
0,
1
] | [
"Dans",
"le",
"traitement",
"du",
"signal",
"numérique",
"et",
"la",
"théorie",
"de",
"l'",
"information",
",",
"la",
"fonction",
"sinc",
"normalisée",
"est",
"généralement",
"définie",
"comme",
"suit"
] | Dans le traitement du signal numérique et la théorie de l'information, la fonction sinc normalisée est généralement définie comme suit | [
[
0,
4
],
[
5,
7
],
[
8,
18
],
[
19,
21
],
[
22,
28
],
[
29,
38
],
[
39,
41
],
[
42,
44
],
[
45,
52
],
[
53,
55
],
[
56,
58
],
[
58,
69
],
[
69,
70
],
[
71,... |
ai-train-98 | [
[
3,
8,
"field"
],
[
10,
11,
"researcher"
],
[
17,
20,
"conference"
],
[
24,
28,
"organisation"
],
[
30,
30,
"organisation"
]
] | [
0,
1,
2,
3,
4
] | [
[
3,
8,
10,
11,
"origin",
"coined_term",
false,
false
],
[
10,
11,
17,
20,
"role",
"",
false,
false
],
[
10,
11,
24,
28,
"role",
"",
false,
false
],
[
30,
30,
24,
28,
"named",
"",
... | [
0,
1,
2,
3
] | [
"Le",
"terme",
"\"",
"linguistique",
"informatique",
"\"",
"a",
"été",
"inventé",
"par",
"David",
"Hays",
",",
"membre",
"fondateur",
"de",
"l'",
"Association",
"for",
"Computational",
"Linguistics",
"et",
"de",
"l'",
"International",
"Committee",
"on",
"Computa... | Le terme "linguistique informatique" a été inventé par David Hays, membre fondateur de l'Association for Computational Linguistics et de l'International Committee on Computational Linguistics (ICCL). | [
[
0,
2
],
[
3,
8
],
[
9,
10
],
[
10,
22
],
[
23,
35
],
[
35,
36
],
[
37,
38
],
[
39,
42
],
[
43,
50
],
[
51,
54
],
[
55,
60
],
[
61,
65
],
[
65,
66
],
[
67,... |
ai-train-99 | [
[
10,
21,
"misc"
],
[
9,
9,
"misc"
],
[
37,
39,
"metrics"
],
[
41,
41,
"metrics"
]
] | [
0,
1,
2,
3
] | [
[
41,
41,
37,
39,
"named",
"",
false,
false
]
] | [
0
] | [
"59",
",",
"pp.",
"2547-2553",
",",
"oct.",
"2011",
"Dans",
"le",
"DPD",
"à",
"mémoire",
"(",
"ou",
"sans",
"mémoire",
")",
"à",
"base",
"de",
"polynômes",
"unidimensionnels",
",",
"afin",
"de",
"résoudre",
"les",
"coefficients",
"des",
"polynômes",
"du",
... | 59, pp. 2547-2553, oct. 2011 Dans le DPD à mémoire (ou sans mémoire) à base de polynômes unidimensionnels, afin de résoudre les coefficients des polynômes du pré-distueur numérique et de minimiser l'erreur quadratique moyenne (EQM), la sortie déformée du système non linéaire doit être sur-échantillonnée à un taux qui p... | [
[
0,
2
],
[
2,
3
],
[
4,
7
],
[
8,
17
],
[
17,
18
],
[
19,
23
],
[
24,
28
],
[
29,
33
],
[
34,
36
],
[
37,
40
],
[
41,
42
],
[
43,
50
],
[
51,
52
],
[
52,
... |
ai-train-100 | [
[
0,
1,
"researcher"
],
[
10,
10,
"location"
],
[
12,
14,
"location"
],
[
16,
17,
"country"
],
[
21,
21,
"location"
],
[
23,
23,
"country"
],
[
35,
43,
"organisation"
],
[
45,
48,
"organisatio... | [
0,
1,
2,
3,
4,
5,
6,
7,
8,
9
] | [
[
0,
1,
10,
10,
"physical",
"",
false,
false
],
[
0,
1,
45,
48,
"physical",
"",
false,
false
],
[
0,
1,
54,
55,
"role",
"",
false,
false
],
[
10,
10,
12,
14,
"physical",
"",
false,... | [
0,
1,
2,
3,
4,
5,
6,
7
] | [
"Boris",
"Katz",
",",
"(",
"né",
"le",
"5",
"octobre",
"1947",
"à",
"Chișinău",
",",
"RSS",
"de",
"Moldavie",
",",
"Union",
"soviétique",
",",
"(",
"aujourd'hui",
"Chișinău",
",",
"Moldavie",
")",
")",
"est",
"un",
"chercheur",
"principal",
"américain",
... | Boris Katz, (né le 5 octobre 1947 à Chișinău, RSS de Moldavie, Union soviétique, (aujourd'hui Chișinău, Moldavie)) est un chercheur principal américain (informaticien) au laboratoire d'informatique et d'intelligence artificielle du MIT au Massachusetts Institute of Technology à Cambridge et chef du groupe InfoLab du la... | [
[
0,
5
],
[
6,
10
],
[
10,
11
],
[
12,
13
],
[
13,
15
],
[
16,
18
],
[
19,
20
],
[
21,
28
],
[
29,
33
],
[
34,
35
],
[
36,
44
],
[
44,
45
],
[
46,
49
],
[
5... |
End of preview. Expand in Data Studio
CrossRE (mirror)
Mirror of the raw CrossRE dataset files (named entity + relation extraction across 6 domains: AI, literature, music, news, politics, science), hosted here for use by DrBenchmark_PARTAGES so collaborators don't need to fetch the data manually.
Original source: Bassignana & Plank, "CrossRE: A Cross-Domain Dataset for Relation Extraction" (Findings of EMNLP 2022). https://github.com/mainlp/CrossRE
Please cite the original paper if you use this data.
- Downloads last month
- 15