This paper addresses the problem of learning an equilibrium efficiently in general-sum Markov games through decentralized multi-agent reinforcement learning. Given fundamental difficulty calculating a Nash (NE), we instead aim at finding coarse correlated (CCE), solution concept that generalizes NE by allowing possible correlations among agents’ strategies. We propose algorithm which each agent...

The general sum-connectivity index of a graph $G$ is defined as $chi_{alpha}(G)= sum_{uvin E(G)} (d_u + d_{v})^{alpha}$ where $d_{u}$ is degree of the vertex $uin V(G)$, $alpha$ is a real number different from $0$ and $uv$ is the edge connecting the vertices $u,v$. In this note, the problem of characterizing the graphs having extremum $chi_{alpha}$ values from a certain collection of polyomino ...

We study the problem of training a principal in multi-agent general-sum game using reinforcement learning (RL). Learning robust policy requires anticipating worst possible strategic responses other agents, which is generally NP-hard. However, we show that no-regret dynamics can identify these worst-case poly-time smooth games. propose framework uses this evaluation method for efficiently RL. Th...