Corpus ID: 202770731

On the Utility of Learning about Humans for Human-AI Coordination

@inproceedings{Carroll2019OnTU,
  title={On the Utility of Learning about Humans for Human-AI Coordination},
  author={Micah Carroll and Rohin Shah and M. Ho and T. Griffiths and S. Seshia and P. Abbeel and A. Dragan},
  booktitle={NeurIPS},
  year={2019}
}
  • Micah Carroll, Rohin Shah, +4 authors A. Dragan
  • Published in NeurIPS 2019
  • Computer Science, Mathematics
  • While we would like agents that can coordinate with humans, current algorithms such as self-play and population-based training create agents that can coordinate with themselves. Agents that assume their partner to be optimal or similar to them can converge to coordination protocols that fail to understand and be understood by humans. To demonstrate this, we introduce a simple environment that requires challenging coordination, based on the popular game Overcooked, and learn a simple model that… CONTINUE READING
    22 Citations

    Figures, Tables, and Topics from this paper

    "Other-Play" for Zero-Shot Coordination
    • 14
    • Highly Influenced
    • PDF
    A Bayesian Framework for Nash Equilibrium Inference in Human-Robot Parallel Play
    • 2
    • PDF
    Learning Personalized Models of Human Behavior in Chess
    • PDF
    Improving Multi-Agent Cooperation using Theory of Mind
    • Highly Influenced
    • PDF
    Watch-And-Help: A Challenge for Social Perception and Human-AI Collaboration
    • PDF
    Open Problems in Cooperative AI
    • PDF
    Learning to Play No-Press Diplomacy with Best Response Policy Iteration
    • 7
    • PDF

    References

    SHOWING 1-10 OF 34 REFERENCES
    Learning Existing Social Conventions via Observationally Augmented Self-Play
    • 15
    Bayesian Action Decoder for Deep Multi-Agent Reinforcement Learning
    • 46
    • PDF
    Planning, Learning and Coordination in Multiagent Decision Processes
    • 421
    • PDF
    Planning for Autonomous Cars that Leverage Effects on Human Actions
    • 230
    • PDF
    Feature-based Joint Planning and Norm Learning in Collaborative Games
    • 6
    • PDF
    Toward Human-Aware Robot Task Planning
    • 79
    • PDF
    On the Utility of Model Learning in HRI
    • 14
    • PDF
    Human-robot cross-training: Computational formulation, modeling and evaluation of a human team training strategy
    • S. Nikolaidis, J. Shah
    • Computer Science
    • 2013 8th ACM/IEEE International Conference on Human-Robot Interaction (HRI)
    • 2013
    • 134
    • PDF
    Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
    • 894
    • PDF
    Shared Autonomy via Hindsight Optimization
    • 103
    • PDF