Interactive Learning

Teaching Imitative Policies in State-space (TIPS)

Reference:

Snehal Jauhri, Carlos E. Celemin, and Jens Kober. Interactive Imitation Learning in State-Space. In Conference on Robot Learning (CoRL), 2020. [bibtex] [pdf] [code] [video]

Interactive Learning of Temporal Features for Control

Reference:

Rodrigo Pérez-Dattari, Carlos E. Celemin, Giovanni Franzese, Javier Ruiz-del-Solar, and Jens Kober. Interactive Learning of Temporal Features for Control: Shaping Policies and State Representations From Human Feedback. IEEE Robotics & Automation Magazine, 27(2):46–54, 2020. [bibtex] [pdf] [doi] [code] [video]

Predictive Probabilistic Merging of Policies (PPMP)

Reference:

Jan Scholten, Daan Wout, Carlos E. Celemin, and Jens Kober. Deep Reinforcement Learning with Feedback-based Exploration. In IEEE Conference on Decision and Control (CDC), pp. 803–808, 2019. [bibtex] [pdf] [doi] [code]

Gaussian Process Coach (GPC)

Reference:

Daan Wout, Jan Scholten, Carlos E. Celemin, and Jens Kober. Learning Gaussian Policies from Corrective Human Feedback. arXiv:1903.05216 [cs.LG], 2019. [bibtex] [pdf] [url] [code]

Enhanced Deep COACH (enhanced D-COACH)

Reference:

Rodrigo Pérez-Dattari, Carlos E. Celemin, Javier Ruiz-del-Solar, and Jens Kober. Continuous Control for High-Dimensional State Spaces: An Interactive Learning Approach. In IEEE International Conference on Robotics and Automation (ICRA), pp. 7611–7617, 2019. [bibtex] [pdf] [doi] [code] [video]

Deep COACH (D-COACH)

Reference:

Rodrigo Pérez-Dattari, Carlos E. Celemin, Javier Ruiz-del-Solar, and Jens Kober. Interactive Learning with Corrective Feedback for Policies based on Deep Neural Networks. In International Symposium on Experimental Robotics (ISER) (Jing Xiao, Torsten Kröger, Oussama Khatib, eds.), Springer International Publishing, pp. 353–363, 2018. [bibtex] [pdf] [doi] [code] [video]

Deep Reinforcement Learning

Fine-tuning Deep RL with Gradient-Free Optimization

Reference:

Tim de Bruin, Jens Kober, Karl Tuyls, and Robert Babuška. Fine-tuning Deep RL with Gradient-Free Optimization. In IFAC World Congress, 2020. [bibtex] [pdf] [code]

Experience Selection Baselines

Reference:

Tim de Bruin, Jens Kober, Karl Tuyls, and Robert Babuška. Experience Selection in Deep Reinforcement Learning for Control. Journal of Machine Learning Research, 19(9):1–56, 2018. [bibtex] [pdf] [html] [code] [video]

Policy Search

Policy learning by Weighting Exploration with the Returns (PoWER)

A basic MATLAB/Octave implementation of Policy learning by Weighting Exploration with the Returns (PoWER) [4 variants: return or state-action value function, constant exploration or automatic adaptation], episodic Reward-Weighted Regression (eRWR), episodic Natural Actor Critic (eNAC), ‘Vanilla’ Policy Gradients (VPG), and Finite Difference Gradients (FDG): matlab_PoWER.zip

The required motor primitive code can be downloaded from http://www-clmc.usc.edu/Resources/Software

Reference:

Jens Kober and Jan Peters. Policy Search for Motor Primitives in Robotics. Machine Learning, 84(1-2):171–203, 2011. [bibtex] [pdf] [doi] [code] [video]

Others

Random Shadows and Highlights

Reference:

Osama Mazhar and Jens Kober. Random Shadows and Highlights: A New Data Augmentation Method for Extreme Lighting Conditions. arXiv:2101.05361 [cs.CV], 2021. [bibtex] [pdf] [url] [code]

DMPs for hitting and batting

A basic MATLAB/Octave implementation: hittingMP.m

Reference:

Jens Kober, Katharina Muelling, Oliver Kroemer, Christoph H. Lampert, Bernhard Schölkopf, and Jan Peters. Movement Templates for Learning of Hitting and Batting. In IEEE International Conference on Robotics and Automation (ICRA), pp. 69–82, 2010. [bibtex] [pdf] [doi] [code] [video]