Safe Policy Learning for Continuous Control

Aleksandra Faust
Aleksandra Faust
Edgar Duenez-Guzman
Edgar Duenez-Guzman
Mohammad Ghavamzadeh
Mohammad Ghavamzadeh

2019.

Cited by: 0|Bibtex|Views7

Abstract:

We study continuous action reinforcement learning problems in which it is crucial that the agent interacts with the environment only through safe policies, ie,~ policies that keep the agent in desirable situations, both during training and at convergence. We formulate these problems as {\em constrained} Markov decision processes (CMDPs) a...More

Code:

Data:

Your rating :
0

 

Tags
Comments