Low Precision Policy Distillation with Application to Low-Power, Real-time Sensation-Cognition-Action Loop with Neuromorphic Computing

Autor:	Mckinstry, Jeffrey L, Barch, Davis R., Bablani, Deepika, Debole, Michael V., Esser, Steven K., Kusnitz, Jeffrey A., Arthur, John V., Modha, Dharmendra S.
Rok vydání:	2018
Předmět:	Computer Science - Machine Learning Computer Science - Neural and Evolutionary Computing Statistics - Machine Learning
Druh dokumentu:	Working Paper
Popis:	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.
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/1809.09260 Zobrazit plný text záznamu View this record from Arxiv