Login / Signup

Dealing with Sparse Rewards in Continuous Control Robotics via Heavy-Tailed Policies.

Souradip ChakrabortyAmrit Singh BediAlec KoppelPratap TokekarDinesh Manocha
Published in: CoRR (2022)
Keyphrases
  • heavy tailed
  • control policy
  • markov decision processes
  • optimal policy
  • computer vision
  • reward function
  • generalized gaussian
  • heavy tails
  • image processing
  • multiscale
  • reinforcement learning
  • latent variables