Sybase>> disout-data-augmentation-procgen>> 返回
项目作者： Nickvanhurck

项目描述 ：
This repository has code that can train and test on Procgen environments. This code includes a PPO baseline, a data augmentation algorithm on PPO, a PPO algorithm that implements Disout and at last a combination of data augmentation with Disout on PPO.
高级语言： Python
项目地址: git://github.com/Nickvanhurck/disout-data-augmentation-procgen.git


new_testing_rewards_1647674445816.pdf
testing_cum_rewards_1647674445839.pdf
training_cum_rewards_1647674445855.pdf
coinrun_testing_rewards_1647674446091.pdf
coinrun_training_rewards_1647674446252.pdf
new_testing_rewards_1647674446272.pdf
testing_cum_rewards_1647674446297.pdf
training_cum_rewards_1647674446323.pdf
heist_testing_rewards_1647674446662.pdf
heist_training_rewards_1647674446722.pdf
new_testing_rewards_1647674446745.pdf
testing_cum_rewards_1647674446778.pdf
training_cum_rewards_1647674446804.pdf
new_testing_rewards_1647674447078.pdf
starpilot_testing_rewards_1647674447094.pdf
starpilot_training_rewards_1647674447345.pdf
testing_cum_rewards_1647674447368.pdf
training_cum_rewards_1647674447388.pdf
bigfish_testing_1647674444942.pdf
bigfish_testing_rewards_1647674445250.pdf
bigfish_training_1647674445266.pdf
bigfish_training_rewards_1647674445513.pdf