Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Inverse Reinforcement Learning through Structured Classification #5

Open
usajpn opened this issue Dec 24, 2019 · 0 comments
Open

Inverse Reinforcement Learning through Structured Classification #5

usajpn opened this issue Dec 24, 2019 · 0 comments
Labels
IRL Inverse Reinforcement Learning

Comments

@usajpn
Copy link
Owner

usajpn commented Dec 24, 2019

Edouard Klein, Matthieu Geist, Bilal Piot, Olivier Pietquin
https://papers.nips.cc/paper/4551-inverse-reinforcement-learning-through-structured-classification.pdf
NIPS 2012

概要

逆強化学習ループの中の強化学習問題を解かないで良いように最初からエキスパートデータを入力として状態における行動のMulti Class Classification問題を解く。Multi Class Classification問題で求めたパラメータをそのまま報酬関数のパラメータとして使用して強化学習をせずに直接報酬関数を求める。

所感

Inner Loopの強化学習を解かないのは画期的だが、
MaxEnt IRLが出て以来、probabilisticなモデルでなく、SVM系列モデルでの論文なので、流行らなかったか?

@usajpn usajpn added the IRL Inverse Reinforcement Learning label Dec 25, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
IRL Inverse Reinforcement Learning
Projects
None yet
Development

No branches or pull requests

1 participant