Counterfactual Multi-Agent Policy Gradients

@inproceedings{Foerster2017CounterfactualMP,
  title={Counterfactual Multi-Agent Policy Gradients},
  author={Jakob N. Foerster and Gregory Farquhar and Triantafyllos Afouras and Nantas Nardelli and Shimon Whiteson},
  booktitle={AAAI},
  year={2017}
}
Cooperative multi-agent systems can be naturally used to model many real world problems, such as network packet routing and the coordination of autonomous vehicles. There is a great need for new reinforcement learning methods that can efficiently learn decentralised policies for such systems. To this end, we propose a new multi-agent actor-critic method called counterfactual multi-agent (COMA) policy gradients. COMA uses a centralised critic to estimate the Q-function and decentralised actors… CONTINUE READING

Similar Papers

Citations

Publications citing this paper.
SHOWING 1-10 OF 131 CITATIONS

Learning Transferable Cooperative Behavior in Multi-Agent Teams

VIEW 4 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Actor-Attention-Critic for Multi-Agent Reinforcement Learning

VIEW 20 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Artificial Neural Networks and Machine Learning – ICANN 2018

  • Lecture Notes in Computer Science
  • 2018
VIEW 10 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Multi-Agent Common Knowledge Reinforcement Learning

VIEW 12 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Multi-agent Deep Reinforcement Learning with Extremely Noisy Observations

VIEW 4 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2017
2019

CITATION STATISTICS

  • 36 Highly Influenced Citations

  • Averaged 43 Citations per year from 2017 through 2019

References

Publications referenced by this paper.
SHOWING 1-10 OF 43 REFERENCES

Optimal Payoff Functions for Members of Collectives

  • Advances in Complex Systems
  • 2001
VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

Actor-critic algorithms

VIEW 7 EXCERPTS
HIGHLY INFLUENTIAL