Reinforcement Learning with Prediction-Based Rewards - 3 views
-
mkisantal on 01 Nov 18Prediction-based method for encouraging reinforcement learning agents to explore their environments through curiosity (reward for unfamiliar states). Learns some games without any extrinsic reward!
-
mkisantal on 01 Nov 18Fun failure case: agent gets stuck in front of TV.
-
dharmeshtailor on 01 Nov 18Not read this article but on a related note: Curiosity and various metrics for it have been explored for some time in robotics (outside of RL) as a framework for exploring (partially) unfamiliar environments. I came across some papers on this topic applied to UAVs when prep'ing for a PhD app. This one (http://www.cim.mcgill.ca/~yogesh/publications/crv2014.pdf) comes to mind - which used a topic modelling approach.