Just another Network site

[1605.07736] Learning Multiagent Communication with Backpropagation

  • The communication protocol between agents is manually specified and not altered during training.
  • Abstract: Many tasks in AI require the collaboration of multiple agents.
  • The model consists of multiple agents and the communication between them is learned alongside their policy.
  • We apply the model to a diverse set of tasks, demonstrating the ability of the agents to learn to communicate amongst themselves, yielding improved performance over non-communicative agents and baselines.
  • In the paper we explore a simple neural model, called CommNN, that uses continuous communication for fully cooperative tasks.

Read the full article, click here.

@soumithchintala: “”Learning Multiagent Communication with Backpropagation”, cool new paper! Multi-agent is hot”

[1605.07736] Learning Multiagent Communication with Backpropagation

Comments are closed, but trackbacks and pingbacks are open.