Machine Learning for Decision Problems, Hierarchical Representations and Controls, Preference Learning
During his PhD thesis, Riad worked with Michèle Sebag and Marc Schoenauer on reducing the expertise requirements of Policy Learning algorithms allowing uninitiated users to teach robots new tasks. They did so by proposing a learning framework (Preference-based Reinforcement Learning) where the user gives binary feedback (better/worse) to trajectories demonstrated by the robot; reducing the role of the user to that of a mere critic. He is expected during his postdoc to focus on the automatic discovery of structure in robot trajectories and to develop (hierarchical) algorithms capable of exploiting it.
Prior to his PhD, he received a diploma in Computer Engineering from Ecole Nationale Superieure d'Informatique (Algiers, Algeria) and an MSc in Artificial Intelligence and Decision from Université Pierre et Marie Curie (Paris, France).
- Reinforcement Learning and Inverse Reinforcement Learning
- Continuous Optimization
- Preference-based Reinforcement Learning
- Akrour, R.; Pajarinen, J.; Neumann, G.; Peters, J. (2019). Projections for Approximate Policy Iteration Algorithms, Proceedings of the International Conference on Machine Learning (ICML).
- Akrour, R.; Abdolmaleki, A.; Abdulsamad, H.; Peters, J.; Neumann, G. (2018). Model-Free Trajectory-based Policy Optimization with Monotonic Improvement, Journal of Machine Learning Research (JMLR).
- Akrour, R.; Atamna, A.; Peters, J. (2021). Convex Optimization with an Interpolation-based Projection and its Application to Deep Learning, Machine Learning (MACH), 110, 8, pp.2267-2289.
- Akrour, R.; Sorokin, D.; Peters, J.; Neumann, G. (2017). Local Bayesian Optimization of Motor Skills, Proceedings of the International Conference on Machine Learning (ICML).
Preference-based Reinforcement Learning
- Wirth, C.; Akrour, R.; FÃ¼rnkranz, J.; Neumann G. (2017). A Survey of Preference-Based Reinforcement Learning Methods, Journal of Machine Learning Research (JMLR).
- Akrour, R.; Schoenauer, M.; Souplet, J.-C.; Sebag, M. (2014). Programming by Feedback, Proceedings of the International Conference on Machine Learning (ICML).