GUNREAL: GPU-accelerated UNsupervised REinforcement and Auxiliary Learning
| Autor: | Takuya Fukagai, Koichi Shirahata, Youri Coppens, Yasumoto Tomita, Atsushi Ike |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
Computer science
business.industry Deep learning Auxiliary Tasks Deep Learning Deep Reinforcement Learning GPU General Engineering 02 engineering and technology 010501 environmental sciences Intelligence artificielle 01 natural sciences Instruction set Acceleration Computer engineering Server 0202 electrical engineering electronic engineering information engineering Task analysis Reinforcement learning 020201 artificial intelligence & image processing Artificial intelligence Reinforcement business Reinforcement learning algorithm 0105 earth and related environmental sciences ComputingMethodologies_COMPUTERGRAPHICS |
| Zdroj: | International Journal of Networking and Computing, 8 (2 Vrije Universiteit Brussel CANDAR |
| Popis: | Recent state-of-the-art deep reinforcement learning algorithms, such as A3C and UNREAL, are designed to train on a single device with only CPU's. Using GPU acceleration for these algorithms results in low GPU utilization, which means the full performance of the GPU is not reached. Motivated by the architecture changes made by the GA3C algorithm, which gave A3C better GPU acceleration, together with the high learning efficiency of the UNREAL algorithm, this paper extends GA3C with the auxiliary tasks from UNREAL to create a deep reinforcement learning algorithm, GUNREAL, with higher learning efficiency and also benefiting from GPU acceleration. We show that our GUNREAL system achieves 3.8 to 9.5 times faster training speed compared to UNREAL and 73% more training efficiency compared to GA3C. info:eu-repo/semantics/published |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|