Skip to content

a3c plus auxiliary tasks value replay and reward prediction

Notifications You must be signed in to change notification settings

404akhan/unreal-implementation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

unreal

UNREAL paper https://arxiv.org/pdf/1611.05397.pdf

Implementation of three Auxiliary Tasks: Pixel Cotrol, Value Function Replay, Reward Prediction built on top of advantage actor-critic

In plots we see that it gives some speed up on learning easy DoomBasic-v0 environment

To see plots with tensorboard run: tensorboard --logdir=logs

Alt text Alt text Alt text

About

a3c plus auxiliary tasks value replay and reward prediction

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages