The Upper Confidence Bounds (UCB) algorithm is a well-known near-optimal strategy for the stochastic multi-armed bandit problem. Its extensions to trees, such as the Upper Confidence Tree (UCT) algorithm, have resulted in good solutions to the problem of Go. This paper introduces DUCT, a distributed algorithm inspired by UCT, for solving Distributed Constraint Optimization Problems (DCOP). Bounds on the solution quality are provided, and experiments show that, compared to existing DCOP approaches, DUCT is able to solve very large problems much more efficiently, or to find significantly higher quality solutions.

Groups of robots are often able to to accomplish missions that no single robot can achieve by themselves. Teamwork is a very important factor in complex, dynamic domains. In heterogeneous teams, robustness and flexibility are increased by the diversity of the robots, each contributing different capabilities. In such heterogeneous Multi-Robot Systems it is reasonable to explicitly take the robots' capabilities into account when determining which one is best suited for a task. In this paper I present a framework that formalizes robots' capabilities and provides a means to estimate their suitability for a task. In highly unpredictable domains, accurate predictions of the outcomes of a robot's actions are virtually impossible. Approximate models and algorithms are required which help to estimate the outcome with highest possible confidence. The proposed architecture can provide estimates of task solution qualities at three levels of confidence: the lowest level only taking the mere existence of capabilities into account, the middle level considering task-specific details with approximate parameters of the capabilities, and the highest confidence level considering more elaborate planning algorithms.

Successive elimination of candidates is often a route to making manipulation intractable to compute. We prove that eliminating candidates does not necessarily increase the computational complexity of manipulation. However, for many voting rules used in practice, the computational complexity increases. For example, it is already known that it is NP-hard to compute how a single voter can manipulate the result of single transferable voting (the elimination version of plurality voting). We show here that it is NP-hard to compute how a single voter can manipulate the result of the elimination version of veto voting, of the closely related Coombs’ rule, and of the elimination versions of a general class of scoring rules.

Peer prediction mechanisms allow the truthful elicitation of private signals (e.g., experiences, or opinions) in regard to a true world state when this ground truth is unobservable. The original peer prediction method is incentive compatible for any number of agents n >= 2, but relies on a common prior, shared by all agents and the mechanism. The Bayesian Truth Serum (BTS) relaxes this assumption. While BTS still assumes that agents share a common prior, this prior need not be known to the mechanism. However, BTS is only incentive compatible for a large enough number of agents, and the particular number of agents required is uncertain because it depends on this private prior. In this paper, we present a robust BTS for the elicitation of binary information which is incentive compatible for every n >= 3, taking advantage of a particularity of the quadratic scoring rule. The robust BTS is the first peer prediction mechanism to provide strict incentive compatibility for every n >= 3 without relying on knowledge of the common prior. Moreover, and in contrast to the original BTS, our mechanism is numerically robust and ex post individually rational.

Two information decay methods are examined that help multi-agent systems cope with dynamic environments. The agents in this simulation have human-like memory and a mechanism to moderate their communications: they forget internally stored information via temporal decay, and they forget distributed information by filtering it as it passes through a communication network. The agents play a foraging game, in which performance depends on communicating facts and requests and on storing facts in internal memory. Parameters of the game and agent models are tuned to human data. Agent groups with moderated communication in small-world networks achieve optimal performance for typical human memory decay values, while non-adaptive agents benefit from stronger memory decay. The decay and filtering strategies interact with the properties of the network graph in ways suggestive of an evolutionary co-optimization between the human cognitive system and an external social structure.

We consider the classic cake cutting problem where one allocates a divisible cake to n participating agents. Among all valid divisions, fairness and efficiency (a.k.a. ~social welfare) are the most critical criteria to satisfy and optimize, respectively. We study computational complexity of computing an efficiency optimal division given the conditions that the allocation satisfies proportional fairness and assigns each agent a connected piece. For linear valuation functions, we give a polynomial time approximation scheme to compute an efficiency optimal allocation. On the other hand, we show that the problem is NP-hard to approximate within a factor of Ω 1/√ n for general piecewise constant functions, and is NP-hard to compute for normalized functions.

This research proposes the use of imitation based learning to build collaborative strategies for a team of agents. Imitation based learning involves learning from an expert by observing her demonstrating a task and then replicating it. This mechanism makes it extremely easy for a knowledge engineer to transfer knowledge to a software agent via human demonstrations. This research aims to apply imitation to learn not only the strategy of an individual agent but also the collaborative strategy of a team of agents to achieve a common goal. The effectiveness of the proposed methodology is being assessed in the domain of RoboCup Soccer Simulation 3D which is a promising platform to address many of the complex real-world problems and offers a truly dynamic, stochastic, and partially-observable environment.

Supply Chain Formation (SCF) is the process of determining the participants in a supply chain, who will exchange what with whom, and the terms of the exchanges. Decentralized SCF appears as a highly intricate task because agents only possess local information and have limited knowledge about the capabilities of other agents. The decentralized SCF problem has been recently cast as an optimization problem that can be efficiently approximated using max-sum loopy belief propagation. Along this direction, in this paper we propose a novel encoding of the problem into a binary factor graph (containing only binary variables) as well as an alternative algorithm. We empirically show that our approach allows to significantly increase scalability, hence allowing to form supply chains in market scenarios with a large number of participants and high competition.

Complexity of voting manipulation is a prominent research topic in computational social choice. The voting manipulation literature usually assumes that the manipulator is only concerned with improving the outcome of the election from her perspective. However, in practice, the manipulator may also be reluctant to lie, i.e., she may have a preference for submitting a vote that does not deviate too much from her true ranking of the candidates. In this paper, we study the complexity of finding a manipulative vote that achieves the manipulator's goal yet is as close as possible to her true preference order. We analyze this problem for three natural notions of closeness, namely, swap distance, footrule distance, and maximum displacement distance, and a variety of voting rules, such as scoring rules, Bucklin, Copeland, and Maximin. For all three distances, we obtain polynomial-time algorithms for all scoring rules and Bucklin and hardness results for Copeland and Maximin.

In extensive-form games with a large number of actions, careful abstraction of the action space is critically important to performance. In this paper we extend previous work on action abstraction using no-limit poker games as our test domains. We show that in such games it is no longer necessary to choose, a priori, one specific range of possible bet sizes. We introduce an algorithm that adjusts the range of bet sizes considered for each bet individually in an iterative fashion. This flexibility results in a substantially improved game value in no-limit Leduc poker. When applied to no-limit Texas Hold'em our algorithm produces an action abstraction that is about one third the size of a state of the art hand-crafted action abstraction, yet has a better overall game value.