Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Recurrent World Models Facilitate Policy Evolution

About

A generative recurrent neural network is quickly trained in an unsupervised manner to model popular reinforcement learning environments through compressed spatio-temporal representations. The world model's extracted features are fed into compact and simple policies trained by evolution, achieving state of the art results in various environments. We also train our agent entirely inside of an environment generated by its own internal world model, and transfer this policy back into the actual environment. Interactive version of paper at https://worldmodels.github.io

David Ha, J\"urgen Schmidhuber• 2018

Related benchmarks

TaskDatasetResultRank
Signal RecoveryLinear-Gaussian distractor environment (train)
Signal R^20.999
15
Signal RecoveryLinear-Gaussian distractor environment (test)
Signal R^20.999
15
Noise RecoveryLinear-Gaussian distractor environment (train)
Noise R^20.491
10
Showing 3 of 3 rows

Other info

Follow for update