Reciprocal Reward Influence Encourages Cooperation From Self-interested Agents

Abstract

Cooperation between self-interested individuals is a widespread phenomenon in the natural world, but remains elusive in interactions between artificially intelligent agents. Instead, naive reinforcement learning algorithms typically converge to Pareto-dominated outcomes in even the simplest of social dilemmas. An emerging literature on opponent shaping has demonstrated the ability to reach prosocial outcomes by influencing the learning of other agents. However, such methods differentiate through the learning step of other agents or optimize for meta-game dynamics, which rely on privileged access to opponents' learning algorithms or exponential sample complexity, respectively. To provide a learning rule-agnostic and sample-efficient alternative, we introduce Reciprocators, reinforcement learning agents which are intrinsically motivated to reciprocate the influence of opponents' actions on their returns. This approach seeks to modify other agents' \(Q\)-values by increasing their return

Stats

• citations0
• S2 citations—
• github stars8
• HF likes0
• heat score1.91
• arxiv keyzhou2024reciprocal

Related papers

• Cooperation And Reputation Dynamics With Reinforcement Learning (2021)3.58
• Evolving Intrinsic Motivations For Altruistic Behavior (2018)2.26
• Multi-agent Cooperation Through Learning-aware Policy Gradients (2024)0.00
• Intrinsic Fluctuations Of Reinforcement Learning Promote Cooperation (2022)9.23
• Balancing Rational And Other-regarding Preferences In Cooperative-competitive Environments (2021)0.00
• Improved Cooperation By Balancing Exploration And Exploitation In Intertemporal Social Dilemma Tasks (2021)0.00
• Cooperative Artificial Intelligence (2022)0.00
• Social Influence As Intrinsic Motivation For Multi-agent Deep Reinforcement Learning (2018)0.00