Skip to content

Sample Complexity of Model-Free Opponent Shaping

Notifications You must be signed in to change notification settings

rmaxmfos/rmaxmfos

Repository files navigation

Implementation of M-FOS with Rmax as the meta-agent

The project is implemented in Python (Rmax-arch1_batch_limitedtraj.py) with the goal of finding the sample complexity of the M-FOS algorithm empirically.

Inner game Environment: Matching Pennies (env_mp_simple.py)

Meta-agent: Rmax (rmax_1_batch_limitedtraj.py)

About

Sample Complexity of Model-Free Opponent Shaping

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published