Global optimality of softmax policy gradient with single hidden layer neural networks in the mean-field regime

10/22/2020
by   Andrea Agazzi, et al.
0

We study the problem of policy optimization for infinite-horizon discounted Markov Decision Processes with softmax policy and nonlinear function approximation trained with policy gradient algorithms. We concentrate on the training dynamics in the mean-field regime, modeling e.g., the behavior of wide single hidden layer neural networks, when exploration is encouraged through entropy regularization. The dynamics of these models is established as a Wasserstein gradient flow of distributions in parameter space. We further prove global optimality of the fixed points of this dynamics under mild conditions on their initialization.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset