Adaptive Punishment for Cooperation in Mixed-Motive Games

• URL: http://arxiv.org/abs/2605.24516v1
• Date: Sat, 23 May 2026 11:01:49 GMT
• Title: Adaptive Punishment for Cooperation in Mixed-Motive Games
• Abstract summary: Self-interested agents often defect for immediate rewards, overlooking the potential of altruistic cooperation to improve long-term gains and collective welfare.<n>Peer punishment can deter defection, but as costly second-order altruism, its persistent imposition may undermine the punisher's interests.<n>We propose Adaptive Punishment for Cooperation, which determines punishment intensity based on both a dynamic punishment probability and the severity of defection.
• Score: 7.763884853916347
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mixed-motive scenarios are ubiquitous in real-world multi-agent interactions, where self-interested agents often defect for immediate rewards, overlooking the potential of altruistic cooperation to improve long-term gains and collective welfare. Peer punishment can deter defection, but as costly second-order altruism, its persistent imposition may undermine the punisher's interests. Existing approaches often struggle to effectively implement punishment to promote cooperation. To balance the efficacy and cost of punishment, we propose Adaptive Punishment for Cooperation (APC), a distributed method that determines punishment intensity based on both a dynamic punishment probability and the severity of defection. This dynamic probability substantially reduces costly and ineffective punishment while also promotes cooperation. To accurately assess defection and its severity, we use a defection awareness module, whose learning is guided by game reward. Theoretical analysis and empirical results show APC performs effectively in iterated public goods game. Empirically, APC also significantly outperforms existing baselines across sequential social dilemmas, learning rational and effective punishment policies that foster cooperation by strategically deterring defection.

Related papers

• Co-evolution of social reward and punishment under institutional interventions [4.789843817744408]
  We investigate how peer and institutional incentives jointly shape the evolution of cooperation, social welfare, and enforcement efficiency in social dilemmas.<n>In a Prisoners Dilemma with four strategies, we allow decentralised peer incentives and centralised institutional incentives to act simultaneously.<n>We find that peer punishment most strongly promotes cooperation, whereas peer reward is more beneficial for social welfare.
  arXiv  Detail & Related papers  (2026-08-02T12:06:46Z)
• Social welfare optimisation under institutional reward and punishment [6.8711834942604915]
  We develop a welfare-centric framework for institutional incentives in finite, well-mixed populations playing a social dilemma.<n>We derive explicit expressions for expected social welfare and how it depends on incentive efficiency and selection intensity.<n>Our results reveal a systematic gap between incentives optimised for cost or cooperation frequency and those that maximise welfare.
  arXiv  Detail & Related papers  (2026-05-29T14:04:55Z)
• Integrated strong reciprocity enables productive punishment and protective defection [1.4323566945483497]
  We analyze an evolutionary game model that integrates upstream and downstream reciprocity with costly punishment.<n>We demonstrate that ISR admits a stable mixed equilibrium of ISR and unconditional defection.<n>At the same time, the mixed equilibrium of ISR and ALLD remains robust under modest complexity.
  arXiv  Detail & Related papers  (2026-01-07T08:03:49Z)
• Rethinking Reward Miscalibration of GRPO in Agentic RL [18.495499496405635]
  We show that outcome based reward ensures expected negative advantage for those flawed middle steps.<n>We propose training the actor to classify good or bad actions to separate the embedding of good/bad actions.
  arXiv  Detail & Related papers  (2025-09-28T13:24:38Z)
• Integrative Experiments Identify How Punishment Impacts Welfare in Public Goods Games [2.359185514776351]
  We study how punishment's impact varies across cooperative settings through a large-scale integrative experiment.<n>We develop models that outperformed human forecasters in predicting punishment outcomes in new experiments.<n>Our results refocus the debate over punishment from whether or not it "works" to the specific conditions under which it does and does not work.
  arXiv  Detail & Related papers  (2025-08-23T22:07:27Z)
• Adversarial Training of Reward Models [74.17196154247964]
  We introduce Adv-RM, a novel adversarial training framework that automatically identifies adversarial examples.<n>By leveraging reinforcement learning, Adv-RM trains a policy to expose vulnerabilities in large state-of-the-art reward models.<n>We demonstrate that Adv-RM significantly outperforms conventional reward training.
  arXiv  Detail & Related papers  (2025-04-08T15:38:25Z)
• Deceptive Sequential Decision-Making via Regularized Policy Optimization [54.38738815697299]
  We model autonomous systems as Markov decision processes, with adversaries using inverse reinforcement learning to recover reward functions.<n>We show three regularization strategies for policy synthesis problems that actively deceive an adversary about a system's reward.<n>We show that diversionary, targeted, and equivocal deception all steer the adversary to false beliefs while still attaining a total accumulated reward that is at least 97% of its optimal, non-deceptive value.
  arXiv  Detail & Related papers  (2025-01-30T23:41:40Z)
• Learning to Balance Altruism and Self-interest Based on Empathy in Mixed-Motive Games [47.8980880888222]
  Multi-agent scenarios often involve mixed motives, demanding altruistic agents capable of self-protection against potential exploitation.<n>We propose LASE Learning to balance Altruism and Self-interest based on Empathy.<n>LASE allocates a portion of its rewards to co-players as gifts, with this allocation adapting dynamically based on the social relationship.
  arXiv  Detail & Related papers  (2024-10-10T12:30:56Z)
• The Benefits of Power Regularization in Cooperative Reinforcement Learning [5.141502818619855]
  We argue that explicitly regularizing the concentration of power in cooperative RL systems can result in systems which are more robust to single agent failure. We present two algorithms for training agents towards this power-regularized objective: Sample Based Power Regularization (SBPR) and Power Regularization via Intrinsic Motivation (PRIM) Our experiments demonstrate that both algorithms successfully balance task reward and power, leading to lower power behavior and avoid catastrophic events.
  arXiv  Detail & Related papers  (2024-06-17T06:10:37Z)
• Incentivized Communication for Federated Bandits [67.4682056391551]
  We introduce an incentivized communication problem for federated bandits, where the server shall motivate clients to share data by providing incentives. We propose the first incentivized communication protocol, namely, Inc-FedUCB, that achieves near-optimal regret with provable communication and incentive cost guarantees.
  arXiv  Detail & Related papers  (2023-09-21T00:59:20Z)
• Mimicking Better by Matching the Approximate Action Distribution [48.95048003354255]
  We introduce MAAD, a novel, sample-efficient on-policy algorithm for Imitation Learning from Observations. We show that it requires considerable fewer interactions to achieve expert performance, outperforming current state-of-the-art on-policy methods.
  arXiv  Detail & Related papers  (2023-06-16T12:43:47Z)
• Collaborative Adversarial Training [82.25340762659991]
  We show that some collaborative examples, nearly perceptually indistinguishable from both adversarial and benign examples, can be utilized to enhance adversarial training. A novel method called collaborative adversarial training (CoAT) is thus proposed to achieve new state-of-the-arts.
  arXiv  Detail & Related papers  (2022-05-23T09:41:41Z)
• Case Study: Predictive Fairness to Reduce Misdemeanor Recidivism Through Social Service Interventions [4.651149317838983]
  The Los Angeles City Attorney's Office created a new Recidivism Reduction and Drug Diversion unit (R2D2) We describe a collaboration with this new unit as a case study for the incorporation of predictive equity into machine learning based decision making.
  arXiv  Detail & Related papers  (2020-01-24T23:52:55Z)
• Cooperative Inverse Reinforcement Learning [64.60722062217417]
  We propose a formal definition of the value alignment problem as cooperative reinforcement learning (CIRL) A CIRL problem is a cooperative, partial-information game with two agents human and robot; both are rewarded according to the human's reward function, but the robot does not initially know what this is. In contrast to classical IRL, where the human is assumed to act optimally in isolation, optimal CIRL solutions produce behaviors such as active teaching, active learning, and communicative actions.
  arXiv  Detail & Related papers  (2016-06-09T22:39:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.