statML-arxiv
Collection
Subsets of the statML-arxiv datasplit • 3 items • Updated
text stringlengths 4.1k 24.5k | arxiv_id stringlengths 10 10 | url stringlengths 32 32 | timestamp stringdate 2020-07-29 02:22:50 2023-03-06 02:13:47 | yymm stringdate 2007-01-01 00:00:00 2302-01-01 00:00:00 | language stringclasses 2
values | Qwen3_token_count int64 4.1k 4.1k | uuid stringlengths 36 36 | start_index int64 0 249k | input_ids list |
|---|---|---|---|---|---|---|---|---|---|
for a fixed sample size $n$. In practical, we first restrict $\tau$ in a reasonable range and select the optimal value then according to the minimal AME principal. After specifying $\tau$, the importance of a sample $({\bf x}_i, y_i)$ can be measured by the corresponding Huber loss $\ell_\tau(y_i - {\bf x}_i^{\top}\... | 2112.06134 | https://arxiv.org/abs/2112.06134 | 2021-12-14 02:17:21 | 2112 | en | 4,096 | 3ce12a7e-ef08-4d6a-9c52-e7d30d45a00b | 4,024 | [
369,
264,
8356,
6077,
1379,
400,
77,
12947,
758,
14976,
11,
220,
582,
1156,
8891,
57960,
29143,
3,
304,
264,
13276,
2088,
323,
3293,
279,
22823,
897,
1221,
4092,
311,
279,
17377,
6769,
36,
12435,
13,
220,
4636,
37838,
57960,
29143,
54... |
the ignorable treatment effect estimate $\tau^{t_1,t_2}_{\textup{igno}}$ as:
\begin{align}
\tau_{\textup{ATE}}^{t_1,t_2} &\triangleq E[Y|do(t_1)]-E[Y|do(t_2)];\\
\tau^{t_1,t_2}_{\textup{igno}} &\triangleq E[Y|t_1]-E[Y|t_2].
\end{align}
Below, we use the result in Corollary \ref{cor:TVD} to bound the average tr... | 2107.04661 | https://arxiv.org/abs/2107.04661 | 2021-07-13 02:02:31 | 2107 | en | 4,096 | bf5a3bb8-b387-400d-b3e2-97a6a92f0fd7 | 2,872 | [
279,
5222,
24715,
6380,
2456,
16045,
57960,
29143,
47822,
83,
62,
16,
13960,
62,
17,
51535,
59,
1318,
454,
90,
622,
78,
3417,
3,
438,
510,
59,
7265,
90,
6612,
532,
262,
1124,
29143,
15159,
59,
1318,
454,
90,
2336,
3417,
47822,
83,
... |
estimate of the uplift. In our case, with $e(\mathbf x) = 1/2$, $Z_1$ simplifies to
\[ Z_1 = \left\{ \begin{array}{cc}
2 & \mathrm{if}~T=1, Y=1, \\
-2 & \mathrm{if}~T=0, Y=1, \\
0 & \mathrm{otherwise.}
\end{array} \right.
\]
Observing $Z_1=2$ and $Z_1=-2$ involve an interesting interpretation ... | 2105.05146 | https://arxiv.org/abs/2105.05146 | 2021-05-12 02:28:25 | 2105 | en | 4,096 | 06c536b2-3651-4346-a755-553ee9958a33 | 85,893 | [
16045,
315,
279,
96447,
13,
758,
1039,
1142,
11,
448,
400,
68,
11520,
10374,
13233,
856,
8,
284,
220,
16,
14,
17,
54876,
400,
57,
62,
16,
3,
15491,
9606,
311,
271,
78045,
1863,
62,
16,
284,
1124,
2359,
59,
90,
1124,
7265,
90,
16... |
$21.91 & 70.52$\pm$21.67 & 1.55 & 69.13$\pm$21.70 & -1.95* & 70.24$\pm$21.97 & 0.05 & 69.43$\pm$22.50 & -2.91*** \\ \hline
VGG & 73.02$\pm$23.05 & 74.21$\pm$22.57 & 0.91 & 71.87$\pm$22.14 & -0.84 & 73.69$\pm$23.86 & 0.53 & 74.42$\pm$22.63 & 1.45 \\ \hline
ResNet& 81.39$\pm$17.19 & 80.87$\pm$18.08 & -0.87 & 78.01$\pm$19... | 2007.15951 | https://arxiv.org/abs/2007.15951 | 2020-08-03 02:11:24 | 2007 | en | 4,096 | 0f6521eb-7993-40e7-8a6a-05598d99f7d9 | 12,600 | [
3,
17,
16,
13,
24,
16,
609,
220,
22,
15,
13,
20,
17,
58736,
5187,
3,
17,
16,
13,
21,
22,
609,
220,
16,
13,
20,
20,
609,
220,
21,
24,
13,
16,
18,
58736,
5187,
3,
17,
16,
13,
22,
15,
609,
481,
16,
13,
24,
20,
9,
609,
2... |
G\}$ where $G = {\cal A}({\mathbb R}^2)\times {\cal A}({\mathbb R})$ is the group of diffeomorphisms of the domain of the image, approximated by local affine planar transformations, and affine transformations of the range space of the image, also known as {\em contrast transformations}. These include small translation... | 2106.13870 | https://arxiv.org/abs/2106.13870 | 2021-06-29 02:03:47 | 2106 | en | 4,096 | 2885fdd8-2666-4504-a7b0-252d989e8316 | 5,076 | [
479,
59,
31716,
1380,
400,
38,
284,
28152,
5416,
362,
92,
2306,
59,
10374,
6066,
431,
92,
61,
17,
10699,
15136,
28152,
5416,
362,
92,
2306,
59,
10374,
6066,
431,
5410,
3,
374,
279,
1874,
315,
1396,
1859,
316,
16347,
13664,
315,
279,... |
$G$ indicates the probability from one webpage to another webpage and is between 0 and 1. The sum of the elements in each row is equal to one. Let $\pi=[\pi_1\cdots \pi_n]$ be a row probability distribution vector, where $\pi_n$ is the $n$-th component of this vector. The row probability distribution vector $\pi$ is a... | 2302.14596 | https://arxiv.org/abs/2302.14596 | 2023-03-01 02:18:27 | 2302 | en | 4,096 | fca81465-2b6e-4c52-8b6c-01788975e0c2 | 4,532 | [
400,
38,
3,
14807,
279,
18927,
504,
825,
44610,
311,
2441,
44610,
323,
374,
1948,
220,
15,
323,
220,
16,
13,
576,
2629,
315,
279,
5424,
304,
1817,
2802,
374,
6144,
311,
825,
13,
6771,
57960,
2493,
5818,
59,
2493,
62,
16,
59,
4385,... |
"16\\} \\\\ \\hline\n\\textbf{Wireless Channel} & AWGN with 5 KHz offset \\\\ \\hline\n\\textbf{Wire(...TRUNCATED) | 2202.11699 | https://arxiv.org/abs/2202.11699 | 2022-02-24 02:24:08 | 2202 | en | 4,096 | 3a90d00d-642b-43d9-b5bc-87e49c9a7726 | 6,982 | [16,21,59,92,24984,1124,71,1056,198,59,1318,13233,90,37845,1717,13434,92,609,33338,25167,448,220,20,(...TRUNCATED) |
"Stiefel1935},\n\twhich is denoted by $\\mathcal{S}_{n,p} = \\{X \\in \\Rnp \\mid X^{\\top} X = I_p\(...TRUNCATED) | 2112.14949 | https://arxiv.org/abs/2112.14949 | 2022-01-03 02:07:23 | 2112 | en | 4,096 | c176f194-3173-4c15-8cbb-707b1854d4dd | 226 | [623,4738,301,16,24,18,20,1583,197,8206,374,3371,9253,553,57960,10374,5416,90,50,51535,77,7237,92,28(...TRUNCATED) |
" to, on average, a very small relative change in the weights of $M$ and $M'$ which can be observed (...TRUNCATED) | 2008.01761 | https://arxiv.org/abs/2008.01761 | 2020-09-22 02:22:52 | 2008 | en | 4,096 | babce15c-a871-4989-90d8-87d8f75fe5a3 | 3,999 | [311,11,389,5461,11,264,1602,2613,8674,2297,304,279,14324,315,400,44,3,323,400,44,44701,892,646,387,(...TRUNCATED) |
"4)--(A8) hold, and the Fr\\'echet trees are $\\alpha$-regular, random-split, honest and symmetr(...TRUNCATED) | 2202.04912 | https://arxiv.org/abs/2202.04912 | 2022-02-11 02:11:15 | 2202 | en | 4,096 | dbcefe3c-2789-4928-935c-0dc07e923180 | 14,646 | [19,29621,7,32,23,8,220,3331,11,220,323,279,2869,10169,4737,295,12408,220,525,57960,7141,3,12,22308,(...TRUNCATED) |
End of preview. Expand in Data Studio
Subset of JackHsieh/statML-arxiv. Each document is exactly 4096 tokens. The train split has exactly twice the number of documents as the test split.
| Split | Documents | Tokens |
|---|---|---|
train |
9728 | 39_845_888 |
test |
4864 | 19_922_944 |
- Downloads last month
- 558