Can you point me to algorithms or code for inverse reinforcement learning?

asked Aug 01 '10 at 10:44

Melipone%20Moody's gravatar image

Melipone Moody
221468

edited Sep 24 '10 at 17:11

Joseph%20Turian's gravatar image

Joseph Turian ♦♦
579051125146


2 Answers:

I just implemented the 'projection' version of Pieter Abbeel's 'Apprenticeship Learning via Inverse Reinforcement Learning' algorithm in Python. I'll be updating it with something more approximate in the near future. It's an exact method, so it must enumerate all Q-states at this point.

answered Apr 10 '11 at 03:17

Daniel%20Duckwoth's gravatar image

Daniel Duckwoth
954222938

edited Apr 10 '11 at 03:18

I'm trying to implement max margin or projection version of Pieter Abbeel's 'Apprenticeship Learning via Inverse Reinforcement Learning' algorithm in Python. Would it be possible for you have a discussion with me? or share the code?

Saurabh [email protected]

(Jun 02 '14 at 16:30) Saurabh_Arora

Here is some code.

answered Sep 20 '10 at 22:22

Umar's gravatar image

Umar
904

Thank you!

(Oct 21 '10 at 13:14) Melipone Moody
Your answer
toggle preview

powered by OSQA

User submitted content is under Creative Commons: Attribution - Share Alike; Other things copyright (C) 2010, MetaOptimize LLC.