Multiagent reinforcement learning in the Iterated Prisoner's Dilemma.

By Tuomas Sandholm, Robert H. Crites. 1996

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Multi-agent Reinforcement Learning in Sequential Social Dilemmas (2017, background): “cooperation can be stabilized, e.g., direct reciprocity [4, 5, 6, 7], indirect reciprocity [8], norm enforcement [9, 10], simple reinforcement learning variants [3], multiagent reinforcement learning [11, 12, 13, 14, 15], spatial structure [16], emotions [17], and social network eects [18, 19]. Howe”
Learning with Opponent-Learning Awareness (2017, methodology): “Sandholm and Crites [39] study the dynamics of independent Q-learning in the IPD under a range of different exploration schedules and function approximators.”
A survey and critique of multiagent deep reinforcement learning (2018, background): “We refer the interested reader to seminal works about convergence in multiagent domains [23,41,42,45,113,165,167,277,295,357,367].”
Learning to compete, coordinate, and cooperate in repeated games using reinforcement learning (2011, background): “Speciﬁcally, M-Qubed uses Sarsa reinforcement learning (Rummery and Niranjan 1994) and recurrent state (Moody et al. 2004, Sandholm and Crites 1996) to estimate its future discounted reward for taking an action from a given state.”
Evolutionary Tournament-Based Comparison of Learning and Non-Learning Algorithms for Iterated Games (2007, background): “Particular games like the Prisoner's Dilemma (PD) have received widespread attention both in game theory (Axelrod 1985, 1987) and in multiagent systems (Sandholm & Crites 1995).”
Optimal Payoff Functions for Members of Collectives (2001, background): “These results also illustrate the substantial superiority of these payoff functions to the perhaps the most natural version of the economics technique of “endogenizing externalities”.”
Inequity aversion improves cooperation in intertemporal social dilemmas (2018, methodology): “oposed inﬂuential models based on behavioral game theory. However, these models have limited applicability since they only generate predictions when the problem can be cast as a matrix game (see e.g. [12, 13]). Here we consider a more realistic video-game setting, like those introduced in the behavi”
High-level reinforcement learning in strategy games (2010, background): “Nevertheless, it has been shown to be effective [8, 12].”
Classes of Multiagent Q-learning Dynamics with epsilon-greedy Exploration (2010, unknown): “This section introduces IQL- and summarizes IGA.”
Coordination and Adaptation in Impromptu Teams (2005, background): “Case-studies on particular games (Sandholm & Crites 1996; Stimpson & Goodrich 2003), such as iterated prisoner’s dilemma, have sometimes dropped this assumption.”
Reaching pareto-optimality in prisoner’s dilemma using conditional joint action learning (2007, methodology): “This gaming environment is different from a repeated Prisoner’s Dilemma problem as dealt by Sandholm et. al ( Sandholm & Crites 1995 ) where agents use the information about duration of the game to compute their expected utility.”
Social Diversity and Social Preferences in Mixed-Motive Reinforcement Learning (2020, background): “The latter category is typified by games like the Prisoner’s Dilemma [46, 53] and the tragedy of the commons [20, 31].”
Learning to cooperate in multi-agent social dilemmas (2006, background): “The desired outcome of our algorithms is to let a MAS of Q-learners collectively learn equilibrium points that payo. dominate the best response dynamics of normal Q-learners.”
Emotional Multiagent Reinforcement Learning in Spatial Social Dilemmas (2015, background): “For example, Sandholm and Crites [20] and Masuda and Ohtsuki [21] defined local state as the sensations of past actions of the interacting agents, while [22]–[24] evolved or reshaped individual rewards during interactions so that agents could rapidly achieve cooperation.”
A Survey of Collectives (2004, background): “In particular, game theory [11, 19, 30, 87], mechanism design [82, 87], and multiagent reinforcement learning [53, 56, 112, 192] are fields that grapple with some of the issues encountered in the field of collectives.”