Goal Misgeneralization in Deep RL

Code for the paper Goal Misgeneralization in Deep Reinforcement Learning, based on the repository https://github.com/joonleesky/train-procgen-pytorch

Requirements

First, install the modified version of the procgen library available here.

Name		Name	Last commit message	Last commit date
Latest commit History 164 Commits
agents		agents
common		common
experiments/scripts		experiments/scripts
hyperparams/procgen		hyperparams/procgen
.gitignore		.gitignore
Readme.md		Readme.md
compute_metrics.py		compute_metrics.py
config.py		config.py
plot_training_csv.py		plot_training_csv.py
plot_value_coin_barchart.py		plot_value_coin_barchart.py
render.py		render.py
run_coinrun.py		run_coinrun.py
run_utils.py		run_utils.py
run_vanilla_coinrun.sh		run_vanilla_coinrun.sh
train-interleave-envs.py		train-interleave-envs.py
train.py		train.py