arXiv subjects

Joseph Y. Halpern

Publications and source records attributed to Joseph Y. Halpern.

At least 127 records · Page 7Linked to original sources

arXiv2011

I Don't Want to Think About it Now:Decision Theory With Costly Computation

Computation plays a major role in decision making. Even if an agent is willing to ascribe a probability to all states and a utility to all outcomes, and maximize expected utility, doing so might present serious computational problems. Moreover, computing the outcome of a given act might be difficult. In a companion paper we develop a framework for game theory with costly computation, where the objects of choice are Turing machines. Here we apply that framework to decision theory. We show how well-known phenomena like first-impression-matters biases (i.e., people tend to put more weight on evidence they hear early on), belief polarization (two people with different prior beliefs, hearing the same evidence, can end up with diametrically opposed conclusions), and the status quo bias (people are much more likely to stick with what they already have) can be easily captured in that framework. Finally, we use the framework to define some new notions: value of computational information (a computational variant of value of information) and and computational value of conversation.

cs.GT↗

arXiv2011

No justified complaints: On fair sharing of multiple resources

Fair allocation has been studied intensively in both economics and computer science, and fair sharing of resources has aroused renewed interest with the advent of virtualization and cloud computing. Prior work has typically focused on mechanisms for fair sharing of a single resource. We provide a new definition for the simultaneous fair allocation of multiple continuously-divisible resources. Roughly speaking, we define fairness as the situation where every user either gets all the resources he wishes for, or else gets at least his entitlement on some bottleneck resource, and therefore cannot complain about not getting more. This definition has the same desirable properties as the recently suggested dominant resource fairness, and also handles the case of multiple bottlenecks. We then prove that a fair allocation according to this definition is guaranteed to exist for any combination of user requests and entitlements (where a user's relative use of the different resources is fixed). The proof, which uses tools from the theory of ordinary differential equations, is constructive and provides a method to compute the allocations numerically.

cs.DC↗

arXiv2010

MDPs with Unawareness

Markov decision processes (MDPs) are widely used for modeling decision-making problems in robotics, automated control, and economics. Traditional MDPs assume that the decision maker (DM) knows all states and actions. However, this may not be true in many situations of interest. We define a new framework, MDPs with unawareness (MDPUs) to deal with the possibilities that a DM may not be aware of all possible actions. We provide a complete characterization of when a DM can learn to play near-optimally in an MDPU, and give an algorithm that learns to play near-optimally when it is possible to do so, as efficiently as possible. In particular, we characterize when a near-optimal solution can be found in polynomial time.

cs.AI↗

arXiv2009

Reasoning About Knowledge of Unawareness Revisited

In earlier work, we proposed a logic that extends the Logic of General Awareness of Fagin and Halpern [1988] by allowing quantification over primitive propositions. This makes it possible to express the fact that an agent knows that there are some facts of which he is unaware. In that logic, it is not possible to model an agent who is uncertain about whether he is aware of all formulas. To overcome this problem, we keep the syntax of the earlier paper, but allow models where, with each world, a possibly different language is associated. We provide a sound and complete axiomatization for this logic and show that, under natural assumptions, the quantifier-free fragment of the logic is characterized by exactly the same axioms as the logic of Heifetz, Meier, and Schipper [2008].

cs.AI↗

arXiv2009

A Logical Characterization of Iterated Admissibility

Brandenburger, Friedenberg, and Keisler provide an epistemic characterization of iterated admissibility (i.e., iterated deletion of weakly dominated strategies) where uncertainty is represented using LPSs (lexicographic probability sequences). Their characterization holds in a rich structure called a complete structure, where all types are possible. Here, a logical charaacterization of iterated admisibility is given that involves only standard probability and holds in all structures, not just complete structures. A stronger notion of strong admissibility is then defined. Roughly speaking, strong admissibility is meant to capture the intuition that "all the agent knows" is that the other agents satisfy the appropriate rationality assumptions. Strong admissibility makes it possible to relate admissibility, canonical structures (as typically considered in completeness proofs in modal logic), complete structures, and the notion of ``all I know''.

cs.AI↗

arXiv2009

Lexicographic probability, conditional probability, and nonstandard probability

The relationship between Popper spaces (conditional probability spaces that satisfy some regularity conditions), lexicographic probability systems (LPS's), and nonstandard probability spaces (NPS's) is considered. If countable additivity is assumed, Popper spaces and a subclass of LPS's are equivalent; without the assumption of countable additivity, the equivalence no longer holds. If the state space is finite, LPS's are equivalent to NPS's. However, if the state space is infinite, NPS's are shown to be more general than LPS's.

cs.GT↗

arXiv2009

Multiagent Learning in Large Anonymous Games

In large systems, it is important for agents to learn to act effectively, but sophisticated multi-agent learning algorithms generally do not scale. An alternative approach is to find restricted classes of games where simple, efficient algorithms converge. It is shown that stage learning efficiently converges to Nash equilibria in large anonymous games if best-reply dynamics converge. Two features are identified that improve convergence. First, rather than making learning more difficult, more agents are actually beneficial in many settings. Second, providing agents with statistical information about the behavior of others can significantly reduce the number of observations needed.

cs.MA↗

arXiv2009

Manipulating Scrip Systems: Sybils and Collusion

Game-theoretic analyses of distributed and peer-to-peer systems typically use the Nash equilibrium solution concept, but this explicitly excludes the possibility of strategic behavior involving more than one agent. We examine the effects of two types of strategic behavior involving more than one agent, sybils and collusion, in the context of scrip systems where agents provide each other with service in exchange for scrip. Sybils make an agent more likely to be chosen to provide service, which generally makes it harder for agents without sybils to earn money and decreases social welfare. Surprisingly, in certain circumstances it is possible for sybils to make all agents better off. While collusion is generally bad, in the context of scrip systems it actually tends to make all agents better off, not merely those who collude. These results also provide insight into the effects of allowing agents to advertise and loan money. While many extensions of Nash equilibrium have been proposed that address collusion and other issues relevant to distributed and peer-to-peer systems, our results show that none of them adequately address the issues raised by sybils and collusion in scrip systems.

cs.GT↗

arXiv2008

Iterated Regret Minimization: A More Realistic Solution Concept

For some well-known games, such as the Traveler's Dilemma or the Centipede Game, traditional game-theoretic solution concepts--and most notably Nash equilibrium--predict outcomes that are not consistent with empirical observations. In this paper, we introduce a new solution concept, iterated regret minimization, which exhibits the same qualitative behavior as that observed in experiments in many games of interest, including Traveler's Dilemma, the Centipede Game, Nash bargaining, and Bertrand competition. As the name suggests, iterated regret minimization involves the iterated deletion of strategies that do not minimize regret.

cs.GT↗

arXiv2008

Toward Expressive and Scalable Sponsored Search Auctions

Internet search results are a growing and highly profitable advertising platform. Search providers auction advertising slots to advertisers on their search result pages. Due to the high volume of searches and the users' low tolerance for search result latency, it is imperative to resolve these auctions fast. Current approaches restrict the expressiveness of bids in order to achieve fast winner determination, which is the problem of allocating slots to advertisers so as to maximize the expected revenue given the advertisers' bids. The goal of our work is to permit more expressive bidding, thus allowing advertisers to achieve complex advertising goals, while still providing fast and scalable techniques for winner determination.

cs.DB↗

arXiv2008

Game Theory with Costly Computation

We develop a general game-theoretic framework for reasoning about strategic agents performing possibly costly computation. In this framework, many traditional game-theoretic results (such as the existence of a Nash equilibrium) no longer hold. Nevertheless, we can use the framework to provide psychologically appealing explanations to observed behavior in well-studied games (such as finitely repeated prisoner's dilemma and rock-paper-scissors). Furthermore, we provide natural conditions on games sufficient to guarantee that equilibria exist. As an application of this framework, we consider a notion of game-theoretic implementation of mediators in computational games. We show that a special case of this notion is equivalent to a variant of the traditional cryptographic definition of protocol security; this result shows that, when taking computation into account, the two approaches used for dealing with "deviating" players in two different communities -- Nash equilibrium in game theory and zero-knowledge "simulation" in cryptography -- are intimately related.

cs.GT↗

arXiv2008

An Almost-Surely Terminating Polynomial Protocol for Asynchronous Byzantine Agreement with Optimal Resilience

Consider an asynchronous system with private channels and $n$ processes, up to $t$ of which may be faulty. We settle a longstanding open question by providing a Byzantine agreement protocol that simultaneously achieves three properties: 1. (optimal) resilience: it works as long as $n>3t$ 2. (almost-sure) termination: with probability one, all nonfaulty processes terminate 3. (polynomial) efficiency: the expected computation time, memory consumption, message size, and number of messages sent are all polynomial in $n$. Earlier protocols have achieved only two of these three properties. In particular, the protocol of Bracha is not polynomially efficient, the protocol of Feldman and Micali is not optimally resilient, and the protocol of Canetti and Rabin does not have almost-sure termination. Our protocol utilizes a new primitive called shunning (asynchronous) verifiable secret sharing (SVSS), which ensures, roughly speaking, that either a secret is successfully shared or a new faulty process is ignored from this point onwards by some nonfaulty process.

cs.DC↗

arXiv2008

A Formal Foundation for XrML

XrML is becoming a popular language in industry for writing software licenses. The semantics for XrML is implicitly given by an algorithm that determines if a permission follows from a set of licenses. We focus on a fragment of the language and use it to highlight some problematic aspects of the algorithm. We then correct the problems, introduce formal semantics, and show that our semantics captures the (corrected) algorithm. Next, we consider the complexity of determining if a permission is implied by a set of XrML licenses. We prove that the general problem is undecidable, but it is polynomial-time computable for an expressive fragment of the language. We extend XrML to capture a wider range of licenses by adding negation to the language. Finally, we discuss the key differences between XrML and MPEG-21, an international standard based on XrML.

cs.CR↗

arXiv2008

Beyond Nash Equilibrium: Solution Concepts for the 21st Century

Nash equilibrium is the most commonly-used notion of equilibrium in game theory. However, it suffers from numerous problems. Some are well known in the game theory community; for example, the Nash equilibrium of repeated prisoner's dilemma is neither normatively nor descriptively reasonable. However, new problems arise when considering Nash equilibrium from a computer science perspective: for example, Nash equilibrium is not robust (it does not tolerate ``faulty'' or ``unexpected'' behavior), it does not deal with coalitions, it does not take computation cost into account, and it does not deal with cases where players are not aware of all aspects of the game. Solution concepts that try to address these shortcomings of Nash equilibrium are discussed.

cs.GT↗

arXiv2008

Defaults and Normality in Causal Structures

A serious defect with the Halpern-Pearl (HP) definition of causality is repaired by combining a theory of causality with a theory of defaults. In addition, it is shown that (despite a claim to the contrary) a cause according to the HP condition need not be a single conjunct. A definition of causality motivated by Wright's NESS test is shown to always hold for a single conjunct. Moreover, conditions that hold for all the examples considered by HP are given that guarantee that causality according to (this version) of the NESS test is equivalent to the HP definition.

cs.AI↗

arXiv2008

The Lotus-Eater Attack

Many protocols for distributed and peer-to-peer systems have the feature that nodes will stop providing service for others once they have received a certain amount of service. Examples include BitTorent's unchoking policy, BAR Gossip's balanced exchanges, and threshold strategies in scrip systems. An attacker can exploit this by providing service in a targeted way to prevent chosen nodes from providing service. While such attacks cannot be prevented, we discuss techniques that can be used to limit the damage they do. These techniques presume that a certain number of processes will follow the recommended protocol, even if they could do better by ``gaming'' the system.

cs.NI↗

arXiv2008

From Qualitative to Quantitative Proofs of Security Properties Using First-Order Conditional Logic

A first-order conditional logic is considered, with semantics given by a variant of epsilon-semantics, where p -> q means that Pr(q | p) approaches 1 super-polynomially --faster than any inverse polynomial. This type of convergence is needed for reasoning about security protocols. A complete axiomatization is provided for this semantics, and it is shown how a qualitative proof of the correctness of a security protocol can be automatically converted to a quantitative proof appropriate for reasoning about concrete security.

cs.CR↗

arXiv2007

Lower Bounds on Implementing Robust and Resilient Mediators

We consider games that have (k,t)-robust equilibria when played with a mediator, where an equilibrium is (k,t)-robust if it tolerates deviations by coalitions of size up to k and deviations by up to $t$ players with unknown utilities. We prove lower bounds that match upper bounds on the ability to implement such mediators using cheap talk (that is, just allowing communication among the players). The bounds depend on (a) the relationship between k, t, and n, the total number of players in the system; (b) whether players know the exact utilities of other players; (c) whether there are broadcast channels or just point-to-point channels; (d) whether cryptography is available; and (e) whether the game has a $k+t)-punishment strategy; that is, a strategy that, if used by all but at most $k+t$ players, guarantees that every player gets a worse outcome than they do with the equilibrium strategy.

cs.GT↗