LoginSignup
0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

More than 5 years have passed since last update.

@Rowing0914(Norio Kosaka)

[Review] Independent Q Learning

Last updated at Posted at 2018-06-19

Profile

Titile: MultiAgent Reinforcement Learning: Independent vs Cooperative Agents
Author: Ming Tan
Published Year: 1993
Link: http://web.media.mit.edu/~cynthiab/Readings/tan-MAS-reinfLearn.pdf

Abstraction

Since the author got inspired by the learning behaviour of human beings, he investigated the multi-agent in reinforcement learning by comparing two generic assumption.

  1. Agents who can cooperate with other agents by sharing information
  2. Agent who cannot cooperate with others

And he found that basically in the cases described below, agents could efficiently learn through the cooperation each other.

  1. Sharing sensation
  2. Sharing episodes
  3. Sharing learned policies

Concept

He wanted to find the answer to the questions below.

  1. Given the same number of RL agents, will cooperative agents outperform independent agents who do not communicate during learning?
  2. What is the price of such cooperation?

Tasks

Basic hunter/prey game
Screen Shot 2018-06-20 at 8.56.25.png

Sensorable area for each agent
Screen Shot 2018-06-20 at 8.56.43.png

Results

Screen Shot 2018-06-20 at 8.58.11.png Screen Shot 2018-06-20 at 8.58.00.png Screen Shot 2018-06-20 at 8.57.48.png Screen Shot 2018-06-20 at 8.57.32.png

Maths

Screen Shot 2018-06-20 at 9.07.05.png

References

Check the slide of 49

0
0
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
Sign upLogin
0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?