id
stringlengths 12
15
| title
stringclasses 100
values | content
stringlengths 0
8.16k
| prechunk_id
stringlengths 0
15
| postchunk_id
stringlengths 0
15
| references
stringclasses 100
values |
|---|---|---|---|---|---|
2403.03504#199
|
Graph Visualization for Blockchain Data
|
CoRR , October 2015.
|
2403.03504#198
|
2403.03504#200
|
2403.03504
|
2403.03504#200
|
Graph Visualization for Blockchain Data
|
[21] Yuhang Zhang, Jun Wang, and Jie Luo.
|
2403.03504#199
|
2403.03504#201
|
2403.03504
|
2403.03504#201
|
Graph Visualization for Blockchain Data
|
Heuristic-based address clustering
in bitcoin.
|
2403.03504#200
|
2403.03504#202
|
2403.03504
|
2403.03504#202
|
Graph Visualization for Blockchain Data
|
IEEE Access , 8:210582–210591, 2020.
|
2403.03504#201
|
2403.03504#203
|
2403.03504
|
2403.03504#203
|
Graph Visualization for Blockchain Data
|
[22] Zibin Zheng, Shaoan Xie, Hong Ning Dai, Xiangping Chen, and Huaimin
Wang.
|
2403.03504#202
|
2403.03504#204
|
2403.03504
|
2403.03504#204
|
Graph Visualization for Blockchain Data
|
Blockchain challenges and opportunities: a survey.
|
2403.03504#203
|
2403.03504#205
|
2403.03504
|
2403.03504#205
|
Graph Visualization for Blockchain Data
|
International
Journal of Web and Grid Services , 14(4):352, 2018.
|
2403.03504#204
|
2403.03504#206
|
2403.03504
|
2403.03504#206
|
Graph Visualization for Blockchain Data
|
10
|
2403.03504#205
|
2403.03504
|
|
1811.00742#0
|
Rationality-proof consensus: extended abstract
|
Rationality-proof consensus: extended
abstract
Jean-Philippe Martin (self) and Eunjin (EJ) Jung (RationalMind)
Abstract
Blockchain systems benefit from lessons in prior art such as fault tolerance, distributed systems,
peer-to-peer systems, and game theory. In this paper we argue that blockchain algorithms
should tolerate both
rational
(self-interested) users and
Byzantine
(malicious) ones, rather than
assuming all non-Byzantine users are
altruistic
and follow the protocols blindly.
|
1811.00742#1
|
1811.00742
|
|
1811.00742#1
|
Rationality-proof consensus: extended abstract
|
Such algorithms
are called
BAR-tolerant
[1].
|
1811.00742#0
|
1811.00742#2
|
1811.00742
|
1811.00742#2
|
Rationality-proof consensus: extended abstract
|
To design a BAR-tolerant system, one can follow these three steps:
clearly define the utility function for the rational users, prove the algorithm is such that there is
no benefit from unilaterally deviating (that is, it's a Byzantine Nash Equilibrium), then prove the
algorithm correct assuming the rational actors follow the protocol.
|
1811.00742#1
|
1811.00742#3
|
1811.00742
|
1811.00742#3
|
Rationality-proof consensus: extended abstract
|
We present an example
attack by rational users: the
gatekeeping
attack, where members of a system selfishly decide to
prevent newcomers from joining.
|
1811.00742#2
|
1811.00742#4
|
1811.00742
|
1811.00742#4
|
Rationality-proof consensus: extended abstract
|
This attack may affect any stake-based system where the
existing members prevent newcomers from making a stake, and essentially form a cartel.
|
1811.00742#3
|
1811.00742#5
|
1811.00742
|
1811.00742#5
|
Rationality-proof consensus: extended abstract
|
We
then sketch a BAR-tolerant consensus protocol for blockchain that can defend against this
attack.
|
1811.00742#4
|
1811.00742#6
|
1811.00742
|
1811.00742#6
|
Rationality-proof consensus: extended abstract
|
It relies on a strict order to decide who gets to propose a new block (so there's no need
to race to solve a crypto puzzle) and it relies on hardware ID tokens to make sure every
computer is only represented at most once as a block proposer to mitigate Sybil attacks.
|
1811.00742#5
|
1811.00742#7
|
1811.00742
|
1811.00742#7
|
Rationality-proof consensus: extended abstract
|
It also
defends against the gatekeeper attack.
|
1811.00742#6
|
1811.00742#8
|
1811.00742
|
1811.00742#8
|
Rationality-proof consensus: extended abstract
|
The BAR-tolerant approach is naturally also applicable
to other blockchain algorithms.
|
1811.00742#7
|
1811.00742#9
|
1811.00742
|
1811.00742#9
|
Rationality-proof consensus: extended abstract
|
Introduction
Blockchain algorithms have captured the world's imagination, and are considered for
applications beyond payment services.
|
1811.00742#8
|
1811.00742#10
|
1811.00742
|
1811.00742#10
|
Rationality-proof consensus: extended abstract
|
As they grow, we want to make sure we design these
algorithms carefully so they can work reliably even at a large scale.
|
1811.00742#9
|
1811.00742#11
|
1811.00742
|
1811.00742#11
|
Rationality-proof consensus: extended abstract
|
Reliability and scalability are
not new challenges in computer science so naturally we draw from prior art including distributed
systems, peer-to-peer, and game theory.
|
1811.00742#10
|
1811.00742#12
|
1811.00742
|
1811.00742#12
|
Rationality-proof consensus: extended abstract
|
The latter especially in the context of incentive
compatibility, a key property for Bitcoin or similar systems where software isn't run by a trusted
central authority: the participants might write a modified version of the software, one that
deviates from the specified algorithm in order to gain some unfair benefit to the user.
|
1811.00742#11
|
1811.00742#13
|
1811.00742
|
1811.00742#13
|
Rationality-proof consensus: extended abstract
|
This
selfish behavior is happening already: some Bitcoin miners skip the validation of transactions
even though this could result in including invalid transactions in the ledger.
|
1811.00742#12
|
1811.00742#14
|
1811.00742
|
1811.00742#14
|
Rationality-proof consensus: extended abstract
|
Research shows that
consensus in Bitcoin will be difficult to achieve when the block reward becomes so small that
the transaction fees are the main source of incoming for miners [6].
|
1811.00742#13
|
1811.00742#15
|
1811.00742
|
1811.00742#15
|
Rationality-proof consensus: extended abstract
|
1
BAR tolerance [1] is the property of protocols that work despite both selfish users (called
rational
) and some number of malicious users (called
Byzantine
).
|
1811.00742#14
|
1811.00742#16
|
1811.00742
|
1811.00742#16
|
Rationality-proof consensus: extended abstract
|
We argue that BAR-tolerance
is a desirable property for blockchain systems.
|
1811.00742#15
|
1811.00742#17
|
1811.00742
|
1811.00742#17
|
Rationality-proof consensus: extended abstract
|
In this paper, we discuss an example attack on
the system by rational users and show how a BAR-tolerant blockchain protocol might defend
against it.
|
1811.00742#16
|
1811.00742#18
|
1811.00742
|
1811.00742#18
|
Rationality-proof consensus: extended abstract
|
We sketch this protocol to show how one would go about designing a BAR-tolerant
algorithm; our protocol is based on the consensus from [1].
|
1811.00742#17
|
1811.00742#19
|
1811.00742
|
1811.00742#19
|
Rationality-proof consensus: extended abstract
|
A
gatekeeping
attack is an example attack by rational users in blockchain systems that requires
a stake (security deposit).
|
1811.00742#18
|
1811.00742#20
|
1811.00742
|
1811.00742#20
|
Rationality-proof consensus: extended abstract
|
More precisely, this attack is applicable to any system that requires a
user to make a deposit before participating in any protocol that gives rewards proportional to the
stake.
|
1811.00742#19
|
1811.00742#21
|
1811.00742
|
1811.00742#21
|
Rationality-proof consensus: extended abstract
|
Many Proof-of-Stake consensus algorithms, including those used in Ethereum Casper
FFG and Tendermint, use deposit-slashing [5] as a solution to the Nothing-at-Stake problem [4],
and require a stake from a user before participating in consensus.
|
1811.00742#20
|
1811.00742#22
|
1811.00742
|
1811.00742#22
|
Rationality-proof consensus: extended abstract
|
For a new user to become a
block producer and earn the block rewards, the new user has to deposit a stake first to a special
address, which is a transaction in the chain’s native currency.
|
1811.00742#21
|
1811.00742#23
|
1811.00742
|
1811.00742#23
|
Rationality-proof consensus: extended abstract
|
For the existing block producers,
the expected block reward they could earn decreases as the newcomer joins.
|
1811.00742#22
|
1811.00742#24
|
1811.00742
|
1811.00742#24
|
Rationality-proof consensus: extended abstract
|
If a block
producer is chosen with a probability proportional to its own stake over the total stake in the
system, then as a new comer joins, the total stake increases and the probability of being elected
decreases.
|
1811.00742#23
|
1811.00742#25
|
1811.00742
|
1811.00742#25
|
Rationality-proof consensus: extended abstract
|
If the block producers vote on candidates and get the reward proportional to its own
stake over the total stake, then as a new comer joins, the total stake increases and the amount
of reward for each block decreases.
|
1811.00742#24
|
1811.00742#26
|
1811.00742
|
1811.00742#26
|
Rationality-proof consensus: extended abstract
|
Thus it is a rational behavior for the existing block
producers to not include the deposit transaction in their blocks, which prevents a newcomer
from becoming a block producer.
|
1811.00742#25
|
1811.00742#27
|
1811.00742
|
1811.00742#27
|
Rationality-proof consensus: extended abstract
|
Related Work
Blockchain systems experience attacks from Byzantine users.
|
1811.00742#26
|
1811.00742#28
|
1811.00742
|
1811.00742#28
|
Rationality-proof consensus: extended abstract
|
Bitcoin Gold suffered from double
spending attack [9]: The hacker with the majority of the computing power was able to deposit
the same Bitcoin Gold to an exchange and also to its own wallet.
|
1811.00742#27
|
1811.00742#29
|
1811.00742
|
1811.00742#29
|
Rationality-proof consensus: extended abstract
|
The attacker may have stolen
$18.6 million from the exchange.
|
1811.00742#28
|
1811.00742#30
|
1811.00742
|
1811.00742#30
|
Rationality-proof consensus: extended abstract
|
Monacoin also experienced a selfish-mining attack that cost an
exchange $90K [7].
|
1811.00742#29
|
1811.00742#31
|
1811.00742
|
1811.00742#31
|
Rationality-proof consensus: extended abstract
|
Software bug could also cause a Byzantine error in the system: overflow in
sum created more bitcoins for block reward than what the protocol dictates and also two
addresses received 92.2 billion bitcoins each [3].
|
1811.00742#30
|
1811.00742#32
|
1811.00742
|
1811.00742#32
|
Rationality-proof consensus: extended abstract
|
While it is less evident than Byzantine attacks, the community has been speculating that the
rational users are the majority of the system and may deviate from a desirable behavior.
|
1811.00742#31
|
1811.00742#33
|
1811.00742
|
1811.00742#33
|
Rationality-proof consensus: extended abstract
|
For
example, some bitcoin miners generated invalid blocks [2]: they would rather spend the time to
check the validity of transactions on solving the hash puzzle.
|
1811.00742#32
|
1811.00742#34
|
1811.00742
|
1811.00742#34
|
Rationality-proof consensus: extended abstract
|
Researchers in [6] showed that it
may be rational to fork the chain when there are not many transactions left, and the transaction
fees form the majority of the reward.
|
1811.00742#33
|
1811.00742#35
|
1811.00742
|
1811.00742#35
|
Rationality-proof consensus: extended abstract
|
2
Approach
We need to tolerate both malicious and self-interested (rational) actors.
|
1811.00742#34
|
1811.00742#36
|
1811.00742
|
1811.00742#36
|
Rationality-proof consensus: extended abstract
|
Since some actors may be malicious, we need to use a Byzantine-tolerant algorithm.
|
1811.00742#35
|
1811.00742#37
|
1811.00742
|
1811.00742#37
|
Rationality-proof consensus: extended abstract
|
These
algorithms can function despite arbitrary behavior from some bounded fraction of the
participants (the exact fraction depends on the model.
|
1811.00742#36
|
1811.00742#38
|
1811.00742
|
1811.00742#38
|
Rationality-proof consensus: extended abstract
|
It's typically 1/3, but can be up to 100% if
sufficiently strong assumptions can be made).
|
1811.00742#37
|
1811.00742#39
|
1811.00742
|
1811.00742#39
|
Rationality-proof consensus: extended abstract
|
The key requirement is to bound the fraction of
malicious actors (otherwise they may be able to evict honest actors, prevent correct transactions
from entering the blockchain, etc.).
|
1811.00742#38
|
1811.00742#40
|
1811.00742
|
1811.00742#40
|
Rationality-proof consensus: extended abstract
|
Instead of using a proof of work (that is, relying on the
assumption that the aggregate computing power bought by the honest actors is larger than that
of the malicious actors), one can use modern hardware features such as the TPM chip present
in most motherboards today.
|
1811.00742#39
|
1811.00742#41
|
1811.00742
|
1811.00742#41
|
Rationality-proof consensus: extended abstract
|
The manufacturer guarantees actors cannot pretend to own more
motherboards than they genuinely do.
|
1811.00742#40
|
1811.00742#42
|
1811.00742
|
1811.00742#42
|
Rationality-proof consensus: extended abstract
|
So long as the aggregate buying power of the honest
actors is larger than that of malicious actors, the majority vote of the system will be in honest
hands - without needing to continually consume vast amounts of power.
|
1811.00742#41
|
1811.00742#43
|
1811.00742
|
1811.00742#43
|
Rationality-proof consensus: extended abstract
|
These hardware tokens
prevent Sybil attacks.
|
1811.00742#42
|
1811.00742#44
|
1811.00742
|
1811.00742#44
|
Rationality-proof consensus: extended abstract
|
Remains only to pick who can add the next block to the blockchain: this
could be done at random, or simply by taking turns.
|
1811.00742#43
|
1811.00742#45
|
1811.00742
|
1811.00742#45
|
Rationality-proof consensus: extended abstract
|
Since some actors can be rational (perhaps everyone who is not malicious), we need an
incentive-compatible algorithm.
|
1811.00742#44
|
1811.00742#46
|
1811.00742
|
1811.00742#46
|
Rationality-proof consensus: extended abstract
|
To tolerate both Byzantine and rational actors we use the BAR
approach [1].
|
1811.00742#45
|
1811.00742#47
|
1811.00742
|
1811.00742#47
|
Rationality-proof consensus: extended abstract
|
The idea is to design a protocol that is a Byzantine Nash Equilibrium.
|
1811.00742#46
|
1811.00742#48
|
1811.00742
|
1811.00742#48
|
Rationality-proof consensus: extended abstract
|
This means
that it is in the best interest of each rational actor to follow the protocol as specified.
|
1811.00742#47
|
1811.00742#49
|
1811.00742
|
1811.00742#49
|
Rationality-proof consensus: extended abstract
|
In this
methodology one first needs to specify what the rational actors will consider
costs
and
benefits
when computing their utility function.
|
1811.00742#48
|
1811.00742#50
|
1811.00742
|
1811.00742#50
|
Rationality-proof consensus: extended abstract
|
A protocol is a Byzantine Nash Equilibrium if rational
actors see no increase in their utility from unilaterally deviating from the protocol.
|
1811.00742#49
|
1811.00742#51
|
1811.00742
|
1811.00742#51
|
Rationality-proof consensus: extended abstract
|
We assume
that they consider the Byzantine actor's worst possible behavior when estimating the utility they
would get from following a particular sequence of actions.
|
1811.00742#50
|
1811.00742#52
|
1811.00742
|
1811.00742#52
|
Rationality-proof consensus: extended abstract
|
To prove a BAR-tolerant protocol
correct, we show
1.
|
1811.00742#51
|
1811.00742#53
|
1811.00742
|
1811.00742#53
|
Rationality-proof consensus: extended abstract
|
That the algorithm is a Byzantine Nash Equilibrium, and then
2.
|
1811.00742#52
|
1811.00742#54
|
1811.00742
|
1811.00742#54
|
Rationality-proof consensus: extended abstract
|
That the algorithm has the desired properties, under the assumption that the rational
actors obey the protocol.
|
1811.00742#53
|
1811.00742#55
|
1811.00742
|
1811.00742#55
|
Rationality-proof consensus: extended abstract
|
The BAR model allows (but does not require) some of the actors to be altruistic, meaning that
they follow the protocol even if it weren't in their personal best interest to do so.
|
1811.00742#54
|
1811.00742#56
|
1811.00742
|
1811.00742#56
|
Rationality-proof consensus: extended abstract
|
In this paper we
do not require (nor take advantage of) altruistic actors.
|
1811.00742#55
|
1811.00742#57
|
1811.00742
|
1811.00742#57
|
Rationality-proof consensus: extended abstract
|
The model also assumes that rational
actors do not collude.
|
1811.00742#56
|
1811.00742#58
|
1811.00742
|
1811.00742#58
|
Rationality-proof consensus: extended abstract
|
This doesn't mean that no collusion happens, it just means that if some
actors collude then they count against the limit on Byzantine actors.
|
1811.00742#57
|
1811.00742#59
|
1811.00742
|
1811.00742#59
|
Rationality-proof consensus: extended abstract
|
When designing a
BAR-tolerant protocol, the author chooses what counts as costs or benefits.
|
1811.00742#58
|
1811.00742#60
|
1811.00742
|
1811.00742#60
|
Rationality-proof consensus: extended abstract
|
This is a trade-off:
including more things as costs or benefits means it applies to more individuals in real life, but it
also makes for a craftier rational actor and the protocol needs to be correspondingly stronger
3
(and its correctness argument gets correspondingly longer).
|
1811.00742#59
|
1811.00742#61
|
1811.00742
|
1811.00742#61
|
Rationality-proof consensus: extended abstract
|
Including only a few things (for
example, perhaps the rational actor is only trying to optimize how many tokens they earn) keeps
the algorithm and proof simpler, but some individuals may not fit this model.
|
1811.00742#60
|
1811.00742#62
|
1811.00742
|
1811.00742#62
|
Rationality-proof consensus: extended abstract
|
This isn't as bad as
it sounds: those actors that act rationally but have different incentives than those the system
was designed to resist will still be tolerated, they will just count against the Byzantine tolerance
threshold.
|
1811.00742#61
|
1811.00742#63
|
1811.00742
|
1811.00742#63
|
Rationality-proof consensus: extended abstract
|
We believe that BAR-tolerance or similar approaches are a good fit for blockchain algorithms
because the high stakes make the temptation to write a modified client harder to resist (so
rational behavior may be seen in practice), and the same stakes also mean that it's important to
rely on as few assumptions as possible (so tolerating arbitrary behavior from some actors is
beneficial).
|
1811.00742#62
|
1811.00742#64
|
1811.00742
|
1811.00742#64
|
Rationality-proof consensus: extended abstract
|
Key benefits
We sketch a blockchain algorithm that can (a) tolerate malicious actors without requiring a
wasteful proof of work, and (b) tolerate rational actors, acting in their individual self-interest.
|
1811.00742#63
|
1811.00742#65
|
1811.00742
|
1811.00742#65
|
Rationality-proof consensus: extended abstract
|
We
hope it serves as an illustration of how one may design a BAR-tolerant algorithm in the context
of a blockchain.
|
1811.00742#64
|
1811.00742#66
|
1811.00742
|
1811.00742#66
|
Rationality-proof consensus: extended abstract
|
Proofs of work in Bitcoin waste large amounts of electricity, creating unnecessary pollution.
|
1811.00742#65
|
1811.00742#67
|
1811.00742
|
1811.00742#67
|
Rationality-proof consensus: extended abstract
|
They defend against
Sybil attacks
, where a single bad actor takes on multiple identities in order
to have more voting power and subvert the protocol.
|
1811.00742#66
|
1811.00742#68
|
1811.00742
|
1811.00742#68
|
Rationality-proof consensus: extended abstract
|
Proof of stake has been proposed as an
alternative; our approach is similar but with small stakes, allowing everyone to join and be
equally rewarded for participating in the chain.
|
1811.00742#67
|
1811.00742#69
|
1811.00742
|
1811.00742#69
|
Rationality-proof consensus: extended abstract
|
Hardware features (e.g.
|
1811.00742#68
|
1811.00742#70
|
1811.00742
|
1811.00742#70
|
Rationality-proof consensus: extended abstract
|
TPM) can be used to
hinder bad actors trying to pretend to be more than one person: they have to buy real hardware
for each voter so it's not as easy as just sending another network message.
|
1811.00742#69
|
1811.00742#71
|
1811.00742
|
1811.00742#71
|
Rationality-proof consensus: extended abstract
|
Another function of Bitcoin's proof of work is to act as a lottery, to pick a leader.
|
1811.00742#70
|
1811.00742#72
|
1811.00742
|
1811.00742#72
|
Rationality-proof consensus: extended abstract
|
Here instead we
use a round-robin approach, taking turn.
|
1811.00742#71
|
1811.00742#73
|
1811.00742
|
1811.00742#73
|
Rationality-proof consensus: extended abstract
|
We rely on a BAR-tolerant protocol to ensure that we
can reach consensus on which block to add to the chain, without giving rational actors room to
deviate for gain.
|
1811.00742#72
|
1811.00742#74
|
1811.00742
|
1811.00742#74
|
Rationality-proof consensus: extended abstract
|
Components
The protocol is built modularly, out of multiple component parts.
|
1811.00742#73
|
1811.00742#75
|
1811.00742
|
1811.00742#75
|
Rationality-proof consensus: extended abstract
|
Going in approximate
chronological order, a user can first read and write transactions in the chain.
|
1811.00742#74
|
1811.00742#76
|
1811.00742
|
1811.00742#76
|
Rationality-proof consensus: extended abstract
|
Then if they want
to produce blocks (and be rewarded for it), they go through a joining and registration step.
|
1811.00742#75
|
1811.00742#77
|
1811.00742
|
1811.00742#77
|
Rationality-proof consensus: extended abstract
|
Then
they follow the consensus subprotocol to wait for their turn and finally produce a block.
|
1811.00742#76
|
1811.00742#78
|
1811.00742
|
1811.00742#78
|
Rationality-proof consensus: extended abstract
|
Keeping the protocol modular helps keep the design simple and the proofs manageable, though
a few of the arguments still need to be over the system as a whole.
|
1811.00742#77
|
1811.00742#79
|
1811.00742
|
1811.00742#79
|
Rationality-proof consensus: extended abstract
|
It also makes it easier to
4
evolve the system over time and to add features.
|
1811.00742#78
|
1811.00742#80
|
1811.00742
|
1811.00742#80
|
Rationality-proof consensus: extended abstract
|
For example in this paper we are not touching
on smart contracts or other high-level concepts, but of course they can also benefit from being
Byzantine and rational tolerant.
|
1811.00742#79
|
1811.00742#81
|
1811.00742
|
1811.00742#81
|
Rationality-proof consensus: extended abstract
|
Joining and Registration
Any user of the system can submit transactions to the blockchain without having to register or
join, but they need to do both in order to produce blocks.
|
1811.00742#80
|
1811.00742#82
|
1811.00742
|
1811.00742#82
|
Rationality-proof consensus: extended abstract
|
Joining is done by submitting a
transaction that transfers a small amount of "coin" to a special address for members (this
address may be managed by a smart contract, or be hardcoded into the consensus protocol
itself).
|
1811.00742#81
|
1811.00742#83
|
1811.00742
|
1811.00742#83
|
Rationality-proof consensus: extended abstract
|
The user can get this deposit back when they're done producing blocks (this may be
done automatically when the user closes the mining program).
|
1811.00742#82
|
1811.00742#84
|
1811.00742
|
1811.00742#84
|
Rationality-proof consensus: extended abstract
|
The deposit is forfeit if the user is
found deviating from the protocol.
|
1811.00742#83
|
1811.00742#85
|
1811.00742
|
1811.00742#85
|
Rationality-proof consensus: extended abstract
|
The registration protocol is part of the same interaction, but with the purpose of making it harder
for a single malicious person to create multiple accounts in an attempt to manipulate voting or
other aspects of the system, i.e.
|
1811.00742#84
|
1811.00742#86
|
1811.00742
|
1811.00742#86
|
Rationality-proof consensus: extended abstract
|
mitigating Sybil attack.
|
1811.00742#85
|
1811.00742#87
|
1811.00742
|
1811.00742#87
|
Rationality-proof consensus: extended abstract
|
Registration can be done by reading a
unique hardware ID from the user's machine, so that each user is only allowed a single identity
in the system.
|
1811.00742#86
|
1811.00742#88
|
1811.00742
|
1811.00742#88
|
Rationality-proof consensus: extended abstract
|
For example one may use the key pair embedded in a TPM, or SGX's linkable
quotes to ensure that a given computer cannot register more than once.
|
1811.00742#87
|
1811.00742#89
|
1811.00742
|
1811.00742#89
|
Rationality-proof consensus: extended abstract
|
Another approach is to
ask the user to pay a one-time fee of some sort.
|
1811.00742#88
|
1811.00742#90
|
1811.00742
|
1811.00742#90
|
Rationality-proof consensus: extended abstract
|
The fee could be a computation (compute
some hashes), human endeavor (mail in a postcard), or paying a token amount.
|
1811.00742#89
|
1811.00742#91
|
1811.00742
|
1811.00742#91
|
Rationality-proof consensus: extended abstract
|
Something that
is small enough that users don't mind doing it once, but that would become a burden if it had to
be done thousands of times.
|
1811.00742#90
|
1811.00742#92
|
1811.00742
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.