Results 1  10
of
109
Online Convex Programming and Generalized Infinitesimal Gradient Ascent
, 2003
"... Convex programming involves a convex set F R and a convex function c : F ! R. The goal of convex programming is to nd a point in F which minimizes c. In this paper, we introduce online convex programming. In online convex programming, the convex set is known in advance, but in each step of some ..."
Abstract

Cited by 288 (4 self)
 Add to MetaCart
(Show Context)
Convex programming involves a convex set F R and a convex function c : F ! R. The goal of convex programming is to nd a point in F which minimizes c. In this paper, we introduce online convex programming. In online convex programming, the convex set is known in advance, but in each step of some repeated optimization problem, one must select a point in F before seeing the cost function for that step. This can be used to model factory production, farm production, and many other industrial optimization problems where one is unaware of the value of the items produced until they have already been constructed. We introduce an algorithm for this domain, apply it to repeated games, and show that it is really a generalization of in nitesimal gradient ascent, and the results here imply that generalized in nitesimal gradient ascent (GIGA) is universally consistent.
Multiagent Learning Using a Variable Learning Rate
 Artificial Intelligence
, 2002
"... Learning to act in a multiagent environment is a difficult problem since the normal definition of an optimal policy no longer applies. The optimal policy at any moment depends on the policies of the other agents and so creates a situation of learning a moving target. Previous learning algorithms hav ..."
Abstract

Cited by 224 (9 self)
 Add to MetaCart
Learning to act in a multiagent environment is a difficult problem since the normal definition of an optimal policy no longer applies. The optimal policy at any moment depends on the policies of the other agents and so creates a situation of learning a moving target. Previous learning algorithms have one of two shortcomings depending on their approach. They either converge to a policy that may not be optimal against the specific opponents' policies, or they may not converge at all. In this article we examine this learning problem in the framework of stochastic games. We look at a number of previous learning algorithms showing how they fail at one of the above criteria. We then contribute a new reinforcement learning technique using a variable learning rate to overcome these shortcomings. Specifically, we introduce the WoLF principle, "Win or Learn Fast", for varying the learning rate. We examine this technique theoretically, proving convergence in selfplay on a restricted class of iterated matrix games. We also present empirical results on a variety of more general stochastic games, in situations of selfplay and otherwise, demonstrating the wide applicability of this method.
Algorithms, Games, and the Internet
 In STOC
, 2001
"... If the Internet is the next great subject for Theoretical Computer Science to model and illuminate mathematically, then Game Theory, and Mathematical Economics more generally, are likely to prove useful tools. In this talk I survey some opportunities and challenges in this important frontier. 1. ..."
Abstract

Cited by 152 (0 self)
 Add to MetaCart
(Show Context)
If the Internet is the next great subject for Theoretical Computer Science to model and illuminate mathematically, then Game Theory, and Mathematical Economics more generally, are likely to prove useful tools. In this talk I survey some opportunities and challenges in this important frontier. 1.
AWESOME: A general multiagent learning algorithm that converges in selfplay and learns a best response against stationary opponents
, 2003
"... A satisfactory multiagent learning algorithm should, at a minimum, learn to play optimally against stationary opponents and converge to a Nash equilibrium in selfplay. The algorithm that has come closest, WoLFIGA, has been proven to have these two properties in 2player 2action repeated games— as ..."
Abstract

Cited by 98 (5 self)
 Add to MetaCart
(Show Context)
A satisfactory multiagent learning algorithm should, at a minimum, learn to play optimally against stationary opponents and converge to a Nash equilibrium in selfplay. The algorithm that has come closest, WoLFIGA, has been proven to have these two properties in 2player 2action repeated games— assuming that the opponent’s (mixed) strategy is observable. In this paper we present AWESOME, the first algorithm that is guaranteed to have these two properties in all repeated (finite) games. It requires only that the other players ’ actual actions (not their strategies) can be observed at each step. It also learns to play optimally against opponents that eventually become stationary. The basic idea behind AWESOME (Adapt When Everybody is Stationary, Otherwise Move to Equilibrium) is to try to adapt to the others’ strategies when they appear stationary, but otherwise to retreat to a precomputed equilibrium strategy. The techniques used to prove the properties of AWESOME are fundamentally different from those used for previous algorithms, and may help in analyzing other multiagent learning algorithms also.
Rational and Convergent Learning in Stochastic Games
, 2001
"... This paper investigates the problem of policy learning in multiagent environments using the stochastic game framework, which we briefly overview. We introduce two properties as desirable for a learning agent when in the presence of other learning agents, namely rationality and convergence. We e ..."
Abstract

Cited by 93 (5 self)
 Add to MetaCart
This paper investigates the problem of policy learning in multiagent environments using the stochastic game framework, which we briefly overview. We introduce two properties as desirable for a learning agent when in the presence of other learning agents, namely rationality and convergence. We examine existing reinforcement learning algorithms according to these two properties and notice that they fail to simultaneously meet both criteria. We then contribute a new learning algorithm, WoLF policy hillclimbing, that is based on a simple principle: "learn quickly while losing, slowly while winning." The algorithm is proven to be rational and we present empirical results for a number of stochastic games showing the algorithm converges.
Convergence and noregret in multiagent learning
 In Advances in Neural Information Processing Systems 17
, 2005
"... Learning in a multiagent system is a challenging problem due to two key factors. First, if other agents are simultaneously learning then the environment is no longer stationary, thus undermining convergence guarantees. Second, learning is often susceptible to deception, where the other agents may be ..."
Abstract

Cited by 85 (0 self)
 Add to MetaCart
(Show Context)
Learning in a multiagent system is a challenging problem due to two key factors. First, if other agents are simultaneously learning then the environment is no longer stationary, thus undermining convergence guarantees. Second, learning is often susceptible to deception, where the other agents may be able to exploit a learner’s particular dynamics. In the worst case, this could result in poorer performance than if the agent was not learning at all. These challenges are identifiable in the two most common evaluation criteria for multiagent learning algorithms: convergence and regret. Algorithms focusing on convergence or regret in isolation are numerous. In this paper, we seek to address both criteria in a single algorithm by introducing GIGAWoLF, a learning algorithm for normalform games. We prove the algorithm guarantees at most zero average regret, while demonstrating the algorithm converges in many situations of selfplay. We prove convergence in a limited setting and give empirical results in a wider variety of situations. These results also suggest a third new learning criterion combining convergence and regret, which we call negative nonconvergence regret (NNR). 1
Analyzing complex strategic interactions in multiagent systems
 In AAAI03 Workshop on Game Theoretic and Decision Theoretic Agents
, 2002
"... We develop a model for analyzing complex games with repeated interactions, for which a full gametheoretic analysis is intractable. Our approach treats exogenously specified, heuristic strategies, rather than the atomic actions, as primitive, and computes a heuristicpayoff table specifying the expe ..."
Abstract

Cited by 75 (3 self)
 Add to MetaCart
We develop a model for analyzing complex games with repeated interactions, for which a full gametheoretic analysis is intractable. Our approach treats exogenously specified, heuristic strategies, rather than the atomic actions, as primitive, and computes a heuristicpayoff table specifying the expected payoffs of the joint heuristic strategy space. We analyze a particular game based on the continuous double auction, and compute Nash equilibria of previously published heuristic strategies. To determine the most plausible equilibria, we study the replicator dynamics of a large population playing the strategies. To account for errors in estimation of payoffs or improvements in strategies, we analyze the dynamics and equilibria based on perturbed payoffs.
New criteria and a new algorithm for learning in multiagent systems
 In Advancesin Neural Information Processing Systems 17
, 2005
"... We propose a new set of criteria for learning algorithms in multiagent systems, one that is more stringent and (we argue) better justified than previous proposed criteria. Our criteria, which apply most straightforwardly in repeated games with average rewards, consist of three requirements: (a) a ..."
Abstract

Cited by 56 (7 self)
 Add to MetaCart
(Show Context)
We propose a new set of criteria for learning algorithms in multiagent systems, one that is more stringent and (we argue) better justified than previous proposed criteria. Our criteria, which apply most straightforwardly in repeated games with average rewards, consist of three requirements: (a) against a specified class of opponents (this class is a parameter of the criterion) the algorithm yield a payoff that approaches the payoff of the best response, (b) against other opponents the algorithm’s payoff at least approach (and possibly exceed) the security level payoff (or maximin value), and (c) subject to these requirements, the algorithm achieve a close to optimal payoff in selfplay. We furthermore require that these average payoffs be achieved quickly. We then present a novel algorithm, and show that it meets these new criteria for a particular parameter class, the class of stationary opponents. Finally, we show that the algorithm is effective not only in theory, but also empirically. Using a recently introduced comprehensive game theoretic test suite, we show that the algorithm almost universally outperforms previous learning algorithms. 1
Learning as search optimization: Approximate large margin methods for structured prediction
 In ICML
, 2005
"... Mappings to structured output spaces (strings, trees, partitions, etc.) are typically learned using extensions of classification algorithms to simple graphical structures (eg., linear chains) in which search and parameter estimation can be performed exactly. Unfortunately, in many complex problems, ..."
Abstract

Cited by 40 (0 self)
 Add to MetaCart
Mappings to structured output spaces (strings, trees, partitions, etc.) are typically learned using extensions of classification algorithms to simple graphical structures (eg., linear chains) in which search and parameter estimation can be performed exactly. Unfortunately, in many complex problems, it is rare that exact search or parameter estimation is tractable. Instead of learning exact models and searching via heuristic means, we embrace this difficulty and treat the structured output problem in terms of approximate search. We present a framework for learning as search optimization, and two parameter updates with convergence theorems and bounds. Empirical evidence shows that our integrated approach to learning and decoding can outperform exact models at smaller computational cost. 1.