No abstract

No abstract

Abstract-We present a view of cooperative control using the language of learning in games. We review the game-theoretic concepts of potential and weakly acyclic games, and demonstrate how several cooperative control problems, such as consensus and dynamic sensor coverage, can be formulated in these settings. Motivated by this connection, we build upon game-theoretic concepts to better accommodate a broader class of cooperative control problems. In particular, we extend existing learning algorithms to accommodate restricted action sets caused by the limitations of agent capabilities and group-based decision making. Furthermore, we also introduce a new class of games called sometimes weakly acyclic games for time-varying objective functions and action sets, and provide distributed algorithms for convergence to an equilibrium.

Log-linear learning is a learning algorithm with equilibrium selection properties. Log-linear learning provides guarantees on the percentage of time that the joint action profile will be at a potential maximizer in potential games. The traditional analysis of log-linear learning has centered around explicitly computing the stationary distribution. This analysis relied on a highly structured setting: i) players' utility functions constitute a potential game, ii) players update their strategies one at a time, which we refer to as asynchrony, iii) at any stage, a player can select any action in the action set, which we refer to as completeness, and iv) each player is endowed with the ability to assess the utility he would have received for any alternative action provided that the actions of all other players remain fixed. Since the appeal of log-linear learning is not solely the explicit form of the stationary distribution, we seek to address to what degree one can relax the structural assumptions while maintaining that only potential function maximizers are the stochastically stable action profiles. In this paper, we introduce slight variants of log-linear learning to include both synchronous updates and incomplete action sets. In both settings, we prove that only potential function maximizers are stochastically stable. Furthermore, we introduce a payoff-based version of log-linear learning, in which players are only aware of the utility they received and the action that they played. Note that log-linear learning in its original form is not a payoff-based learning algorithm. In payoff-based log-linear learning, we also prove that only potential maximizers are stochastically stable. The key enabler for these results is to change the focus of the analysis away from deriving the explicit form of the stationary distribution of the learning process towards characterizing the stochastically stable states. The resulting analysis uses the theory of resistance trees for regular perturbed Markov decision processes, thereby allowing a relaxation of the aforementioned structural assumptions.

We consider a continuous-time form of repeated matrix games in which player strategies evolve in reaction to opponent actions. Players observe each other's actions, but do not have access to other player utilities. Strategy evolution may be of the best response sort, as in fictitious play, or a gradient update. Such mechanisms are known to not necessarily converge. We introduce a form of "dynamic" fictitious and gradient play strategy update mechanisms. These mechanisms use derivative action in processing opponent actions and, in some cases, can lead to behavior converging to Nash equilibria in previously nonconvergent situations. We analyze convergence in the case of exact and approximate derivative measurements of the dynamic update mechanisms. In the ideal case of exact derivative measurements, we show that convergence to Nash equilibrium can always be achieved. In the case of approximate derivative measurements, we derive a characterization of local convergence that shows how the dynamic update mechanisms can converge if the traditional static counterparts do not. We primarily discuss two player games, but also outline extensions to multiplayer games. We illustrate these methods with convergent simulations of the well known Shapley and Jordan counterexamples.