Trust the Model When It Is Confident: Masked Model-based Actor-Critic

Feiyang Pan; Jia He; Dandan Tu; Qing He

NeurIPS 2020poster61 citations

Trust the Model When It Is Confident: Masked Model-based Actor-Critic

Abstract

It is a popular belief that model-based Reinforcement Learning (RL) is more sample efficient than model-free RL, but in practice, it is not always true due to overweighed model errors. In complex and noisy settings, model-based RL tends to have trouble using the model if it does not know when to trust the model.

BibTeX

@inproceedings{NEURIPS2020_77133be2,
 author = {Pan, Feiyang and He, Jia and Tu, Dandan and He, Qing},
 booktitle = {Advances in Neural Information Processing Systems},
 editor = {H. Larochelle and M. Ranzato and R. Hadsell and M.F. Balcan and H. Lin},
 pages = {10537--10546},
 publisher = {Curran Associates, Inc.},
 title = {Trust the Model When It Is Confident: Masked Model-based Actor-Critic},
 url = {https://proceedings.neurips.cc/paper_files/paper/2020/file/77133be2e96a577bd4794928976d2ae2-Paper.pdf},
 volume = {33},
 year = {2020}
}