Skip to content

HumanCompatibleAI/rlsp

Repository files navigation

Reward Learning by Simulating the Past

This is the code accompanying the paper "Preferences Implicit in the State of the World". Paper, blog post, poster.

Tests can be run with python setup.py test.

Instructions for running the experiments can be found in experiments.sh. The script experiments-for-plots.sh generates the plots from the paper.

About

Reward Learning by Simulating the Past

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages