Optimal Control and AI

May 27, 2018

–RL System Image–

Code

  • Torchkit: A toolkit for RL research built using PyTorch.
  • MPG: Platform2D environment and the angular policy gradient algorithm.
Carson Eisenach
Carson Eisenach
Senior Applied Scientist, Amazon

My research interests include optimization and machine learning.

Posts

Torchkit: A toolkit for RL research built using PyTorch
Torchkit v0 now available on Github.
Feb 17, 2019 1 min read
Project
Torchkit: A toolkit for RL research built using PyTorch

Publications

Marginal Policy Gradients: A Unified Family of Estimators for Bounded Action Spaces with Applications
Many complex domains, such as robotics control and real-time strategy (RTS) games, require an agent to learn a continuous control. In …
Preprint Code Project
Marginal Policy Gradients: A Unified Family of Estimators for Bounded Action Spaces with Applications
Natural Policy Gradient for Exponential Families
Recent work has highlighted how a misalignment between the support of the policy and the action space of the reinforcement learning …
PDF Code Project