/inverse_rl

Good intention is all you need!

Primary LanguageJupyter Notebook

Environments

Grid World: https://github.com/yrevar/inverse_rl/blob/master/environments/MDPGridWorld_Experiment.ipynb
Navigation World: https://github.com/yrevar/InverseRL/tree/master/Environments/NavigationGridWorld
Satellite Grid World: https://github.com/yrevar/InverseRL/tree/master/Environments/SatelliteGridWorld

Inverse RL

MaxEntropy IRL: https://github.com/yrevar/InverseRL/blob/master/MaxEntIRL/MaxEntIRL.py
Maximum Likelihood IRL: https://github.com/yrevar/InverseRL/blob/master/MLIRL/MLIRL.py
Receding Horizon IRL (Tensorflow): https://github.com/yrevar/inverse_rl/blob/master/irl/rhirl/RHIRL_Navigation_Experiment.ipynb

References

  1. Ziebart, Brian D., et al. "Maximum Entropy Inverse Reinforcement Learning." AAAI. Vol. 8. 2008.
  2. Matthew Alger's MaxEnt implementation. https://github.com/MatthewJA/Inverse-Reinforcement-Learning
  3. Vroman, Monica C. Maximum likelihood inverse reinforcement learning. Rutgers The State University of New Jersey-New Brunswick, 2014.
  4. MacGlashan, James, and Michael L. Littman. "Between Imitation and Intention Learning." IJCAI. 2015.