Search Results for author: Kagan Tumer

Found 4 papers, 2 papers with code

Evolution-Guided Policy Gradient in Reinforcement Learning

6 code implementations NeurIPS 2018 Shauharda Khadka, Kagan Tumer

However, these methods typically suffer from three core difficulties: temporal credit assignment with sparse rewards, lack of effective exploration, and brittle convergence properties that are extremely sensitive to hyperparameters.

Continuous Control Evolutionary Algorithms +2

Cannot find the paper you are looking for? You can Submit a new open access paper.