Timezone: »

Gossip-based Actor-Learner Architectures for Deep Reinforcement Learning
Mahmoud Assran · Joshua Romoff · Nicolas Ballas · Joelle Pineau · Mike Rabbat

Thu Dec 12 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #230

Multi-simulator training has contributed to the recent success of Deep Reinforcement Learning (Deep RL) by stabilizing learning and allowing for higher training throughputs. In this work, we propose Gossip-based Actor-Learner Architectures (GALA) where several actor-learners (such as A2C agents) are organized in a peer-to-peer communication topology, and exchange information through asynchronous gossip in order to take advantage of a large number of distributed simulators. We prove that GALA agents remain within an epsilon-ball of one-another during training when using loosely coupled asynchronous communication. By reducing the amount of synchronization between agents, GALA is more computationally efficient and scalable compared to A2C, its fully-synchronous counterpart. GALA also outperforms A2C, being more robust and sample efficient. We show that we can run several loosely coupled GALA agents in parallel on a single GPU and achieve significantly higher hardware utilization and frame-rates than vanilla A2C at comparable power draws.

Author Information

Mahmoud Assran (McGill University / Facebook AI Research)
Joshua Romoff (McGill University)
Nicolas Ballas (Facebook FAIR)
Joelle Pineau (Facebook)
Mike Rabbat (Facebook FAIR)

More from the Same Authors