Projected Natural Actor-Critic Philip S. Thomas, William Dabney, Sridhar Mahadevan, and Stephen Giguere School of Computer Science University of Massachusetts Amherst Amherst, MA 01003

First Page		Document Content
Date: 2013-11-10 12:06:12 Numerical analysis Numerical linear algebra Gradient descent Mathematical optimization Gradient method Reinforcement learning Artificial neural network Subgradient method		Projected Natural Actor-Critic Philip S. Thomas, William Dabney, Sridhar Mahadevan, and Stephen Giguere School of Computer Science University of Massachusetts Amherst Amherst, MA 01003 Add to Reading List Source URL: psthomas.com Download Document from Source Website File Size: 1,13 MB Share Document on Facebook