How to Run

This repository contains the code for paper REPRODUCING: UCB EXPLORATION VIA Q-ENSEMBLES.

Clone this repository:

git clone https://github.com/yifjiang/UCB-review.git
cd UCB-review

You may also need to install openai-gym and openai-baseline.

To run the UCB-ensemble algorithm, type:

python run_atari_UCB_Ensemble.py [--env $(AnAtariEnvironmentNoFrameSkip-v0)]

To run the Double-DQN algorithm, type:

python run_atari_double.py [--env $(AnAtariEnvironmentNoFrameSkip-v0)]

Name		Name	Last commit message	Last commit date
Latest commit History 65 Commits
UCB_Ensemble		UCB_Ensemble
double_DQN		double_DQN
my_UpNDown		my_UpNDown
my_baselines		my_baselines
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
Reproducing UCB EXPLORATION VIA Q-ENSEMBLES.pdf		Reproducing UCB EXPLORATION VIA Q-ENSEMBLES.pdf
plot.py		plot.py
run_atari_UCB_Ensemble.py		run_atari_UCB_Ensemble.py
run_atari_UCB_Ensemble_nocnn.py		run_atari_UCB_Ensemble_nocnn.py
run_atari_double.py		run_atari_double.py
run_atari_double_nocnn.py		run_atari_double_nocnn.py
test.py		test.py
train_cartpole.py		train_cartpole.py

Provide feedback