R
Robert Babuska
Researcher at Delft University of Technology
Publications - 381
Citations - 17611
Robert Babuska is an academic researcher from Delft University of Technology. The author has contributed to research in topics: Fuzzy logic & Reinforcement learning. The author has an hindex of 56, co-authored 371 publications receiving 15388 citations. Previous affiliations of Robert Babuska include Carnegie Mellon University & Czech Technical University in Prague.
Papers
More filters
Book ChapterDOI
Least-squares methods for policy iteration
Lucian Busoniu,Alessandro Lazaric,Mohammad Ghavamzadeh,Rémi Munos,Robert Babuska,Bart De Schutter +5 more
TL;DR: This chapter reviews least-squares methods for policy iteration, an important class of algorithms for approximate reinforcement learning, and provides guarantees on the performance obtained asymptotically, as the number of samples processed and iterations executed grows to infinity.
Journal ArticleDOI
Particle Swarms in Optimization and Control
TL;DR: The flexibility, scalability, and robustness to errors on a local level are intrinsic properties of swarms that have attracted the interest of researchers in applying swarm technology to various problems.
Journal ArticleDOI
Modeling and Control of Legged Locomotion via Switching Max-Plus Models
TL;DR: The framework presented in this paper relies on a compact representation of the gait space, provides guarantees regarding the transient and steady-state behavior, and results in simple implementations on legged robotic platforms.
Proceedings ArticleDOI
Optimistic planning for continuous-action deterministic systems
TL;DR: A novel planning algorithm called SOOP is introduced that works for deterministic systems with continuous states and actions, and is the first method to explore the true solution space, consisting of infinite sequences of continuous actions, without requiring knowledge about the smoothness of the system.
Journal ArticleDOI
Complexity reduction in fuzzy modeling
TL;DR: Methods based on similarity analysis that, without performing additional knowledge or data acquisition, allow for the generation of fuzzy models of varying complexity are discussed.