Low Precision Policy Distillation with Application to Low-Power, Real-time Sensation-Cognition-Action Loop with Neuromorphic Computing
This enables efficient, real-time RL on resource-constrained devices, though it is incremental as it adapts existing distillation methods to low-precision settings.
The paper tackled the problem of training low-precision reinforcement learning agents by introducing policy distillation from high-precision networks, achieving real-time end-to-end game playing on low-power neuromorphic hardware across 10 ATARI games.
Low precision networks in the reinforcement learning (RL) setting are relatively unexplored because of the limitations of binary activations for function approximation. Here, in the discrete action ATARI domain, we demonstrate, for the first time, that low precision policy distillation from a high precision network provides a principled, practical way to train an RL agent. As an application, on 10 different ATARI games, we demonstrate real-time end-to-end game playing on low-power neuromorphic hardware by converting a sequence of game frames into discrete actions.