CoBERL: Contrastive BERT for Reinforcement Learning

Autor:	Banino, Andrea, Badia, Adrià Puidomenech, Walker, Jacob, Scholtes, Tim, Mitrovic, Jovana, Blundell, Charles
Rok vydání:	2021
Předmět:	Computer Science - Machine Learning Computer Science - Artificial Intelligence
Druh dokumentu:	Working Paper
Popis:	Many reinforcement learning (RL) agents require a large amount of experience to solve tasks. We propose Contrastive BERT for RL (CoBERL), an agent that combines a new contrastive loss and a hybrid LSTM-transformer architecture to tackle the challenge of improving data efficiency. CoBERL enables efficient, robust learning from pixels across a wide range of domains. We use bidirectional masked prediction in combination with a generalization of recent contrastive methods to learn better representations for transformers in RL, without the need of hand engineered data augmentations. We find that CoBERL consistently improves performance across the full Atari suite, a set of control tasks and a challenging 3D environment. Comment: 9 pages, 2 figures, 6 tables
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2107.05431 Zobrazit plný text záznamu View this record from Arxiv