Parameter Space Noise for Exploration

Matthias Plappert
Matthias Plappert
Prafulla Dhariwal
Prafulla Dhariwal
Szymon Sidor
Szymon Sidor

international conference on learning representations, 2018.

Cited by: 332|Views149
EI

Abstract:

Deep reinforcement learning (RL) methods generally engage in exploratory behavior through noise injection in the action space. An alternative is to add noise directly to the agentu0027s parameters, which can lead to more consistent exploration and a richer set of behaviors. Methods such as evolutionary strategies use parameter perturbatio...More

Code:

Data:

Your rating :
0

 

Tags
Comments