[1602.01783] Asynchronous Methods for Deep Reinforcement Learning

We propose a conceptually simple and lightweight framework for deep reinforcement learning that uses asynchronous gradient descent for optimization of deep neural network controllers. We present asynchronous variants of four standard reinforcement learning algorithms and show that parallel actor-learners have a stabilizing effect on training allowing all four methods to successfully train neural network controllers. The best performing method, an asynchronous variant of actor-critic, surpasses the current state-of-the-art on the Atari domain while training for half the time on a single multi-core CPU instead of a GPU. Furthermore, we show that asynchronous actor-critic succeeds on a wide variety of continuous motor control problems as well as on a new task of navigating random 3D mazes using a visual input.

We propose a conceptually simple and lightweight framework for deep reinforcement learning that uses asynchronous gradient descent for optimization of deep neural network controllers. We present asynchronous variants of four standard reinforcement learning algorithms and show that parallel actor-learners have a stabilizing effect on training allowing all four methods to successfully train neural network controllers. The best performing method, an asynchronous variant of actor-critic, surpasses the current state-of-the-art on the Atari domain while training for half the time on a single multi-c

related reading

• Is Frontier Asynchronous RL Solved? — Luke J. Huangluk-huang.github.io
• A (Long) Peek into Reinforcement Learning | Lil'Loglilianweng.github.io
• Key Papers in Deep RL - Spinning Up documentationspinningup.openai.com
• [1709.06560] Deep Reinforcement Learning that Mattersarxiv.org
• Learning Beyond Gradientstrinkle23897.github.io
• NL.pdfabehrouz.github.io
• The Decade of Deep Learning | Leo Gaobmk.sh
• Loss of plasticity in deep continual learning | Naturenature.com
• Deep Q-Networks Explained — LessWronglesswrong.com
• [1801.01290] Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorarxiv.org
• [1707.06347] Proximal Policy Optimization Algorithmsarxiv.org
• Deep Reinforcement Learning Doesn't Work Yetalexirpan.com