mlsanityData to 7 Oct 2026

Learning Altruistic Behaviours in Reinforcement Learning without External Rewards

arXiv 2107.09598 ↗First posted 20 Jul 2021cs.AI cs.LG cs.MAcited by 8

Not on the map. Why?

Abstract (arXiv)

Can artificial agents learn to assist others in achieving their goals without knowing what those goals are? Generic reinforcement learning agents could be trained to behave altruistically towards others by rewarding them for altruistic behaviour, i.e., rewarding them for benefiting other agents in a given situation. Such an approach assumes that other agents' goals are known so that the altruistic agent can cooperate in achieving those goals. However, explicit knowledge of other agents' goals is often difficult to acquire. In the case of human agents, their goals and preferences may be difficult to express fully; they might be ambiguous or even contradictory. Thus, it is beneficial to develop agents that do not depend on external supervision and learn altruistic behaviour in a task-agnostic manner. We propose to act altruistically towards other agents by giving them more choice and allowing them to achieve their goals better. Some concrete examples include opening a door for others or safeguarding them to pursue their objectives without interference. We formalize this concept and propose an altruistic agent that learns to increase the choices another agent has by preferring to maximize the number of states that the other agent can reach in its future. We evaluate our approach in three different multi-agent environments where another agent's success depends on altruistic behaviour. Finally, we show that our unsupervised agents can perform comparably to agents explicitly trained to work cooperatively, in some cases even outperforming them.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 38 of its 72 references that are in the map. First posted on arXiv 20 Jul 2021.

Builds on

38 of 72 references.

References · 38

  1. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  2. 2015Continuous control with deep reinforcement learningcited by 3,279
  3. 2015Trust Region Policy Optimizationcited by 2,503
  4. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,704
  5. 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,519
  6. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,179
  7. 2017Learning Robust Rewards with Adversarial Inverse Reinforcement Learningcited by 485
  8. 2020Artificial Intelligence, Values and Alignmentcited by 443
  9. 2016Cooperative Inverse Reinforcement Learningcited by 298
  10. 2018A Survey of Inverse Reinforcement Learning: Challenges, Methods and Progresscited by 294
  11. 2016Variational Intrinsic Controlcited by 262
  12. 2017Emergence of Grounded Compositional Language in Multi-Agent Populationscited by 257
26 more
  1. 2018Provably Efficient Maximum Entropy Explorationcited by 240
  2. 2020Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observationscited by 235
  3. 2017Learning with Opponent-Learning Awarenesscited by 218
  4. 2020PettingZoo: Gym for Multi-Agent Reinforcement Learningcited by 209
  5. 2019Efficient Exploration via State Marginal Matchingcited by 186
  6. 2015Variational Information Maximisation for Intrinsically Motivated Reinforcement Learningcited by 142
  7. 2015Taming the Noise in Reinforcement Learning via Soft Updatescited by 125
  8. 2019Optimal Policies Tend to Seek Powercited by 115
  9. 2020Shared Experience Actor-Critic for Multi-Agent Reinforcement Learningcited by 114
  10. 2021State Entropy Maximization with Random Encoders for Efficient Explorationcited by 107
  11. 2018Multi-Agent Generative Adversarial Imitation Learningcited by 99
  12. 2019Multi-Agent Adversarial Inverse Reinforcement Learningcited by 60
  13. 2018Learning Policy Representations in Multiagent Systemscited by 48
  14. 2015Shared Autonomy via Hindsight Optimizationcited by 46
  15. 2017Pragmatic-Pedagogic Value Alignmentcited by 36
  16. 2018Evolving intrinsic motivations for altruistic behaviorcited by 30
  17. 2020Too many cooks: Bayesian inference for coordinating multi-agent collaborationcited by 29
  18. 2019Active Exploration in Markov Decision Processescited by 27
  19. 2020AvE: Assistance via Empowermentcited by 27
  20. 2017A Deep Policy Inference Q-Network for Multi-Agent Systemscited by 24
  21. 2020Task-Agnostic Exploration via Policy Gradient of a Non-Parametric State Entropy Estimatecited by 21
  22. 2018Modelling the Dynamic Joint Policy of Teammates with Attention Multi-agent DDPGcited by 20
  23. 2019An Intrinsically-Motivated Approach for Learning Highly Exploring and Fast Mixing Policiescited by 20
  24. 2020Efficient Empowerment Estimation for Unsupervised Stabilizationcited by 7
  25. 2020Scalable Multi-Agent Inverse Reinforcement Learning via Actor-Attention-Criticcited by 6
  26. 2018Generating Plans that Predict Themselvescited by 5

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 4; 2025: 2.

8 papers.

  1. 2024Beyond Joint Demonstrations: Personalized Expert Guidance for Efficient Multi-Agent Reinforcement Learningcited by 5
  2. 2023Intent-aligned AI systems deplete human agency: the need for agency foundations research in AI safetycited by 4
  3. 2024Toward Finding Strong Pareto Optimal Policies in Multi-Agent Reinforcement Learningcited by 3
  4. 2024Combining Theory of Mind and Kindness for Self-Supervised Human-AI Alignmentcited by 1
  5. 2023A Case for AI Safety via Lawcited by 0
  6. 2024We Urgently Need Intrinsically Kind Machinescited by 0
  7. 2025Understanding Action Effects through Instrumental Empowerment in Multi-Agent Reinforcement Learningcited by 0
  8. 2025When Assisting One Disempowers Anothercited by 0