Mykel J. Kochenderfer
Stanford University
499 Papers
1.6K Citations
Mykel J. Kochenderfer is an academic researcher from Stanford University. The author has contributed to research in topics: Computer science & Markov decision process. The author has an hindex of 41, co-authored 388 publications. Previous affiliations of Mykel J. Kochenderfer include Massachusetts Institute of Technology & University of Edinburgh.
Chat about Author
Papers
Learning an Urban Air Mobility Encounter Model from Expert Preferences
Sydney M. Katz,Anne-Claire Le Bihan,Mykel J. Kochenderfer +2 more
- 01 Sep 2019
TL;DR: In this paper, preference-based learning is extended to tune an encounter model from expert preferences in the form of a stochastic policy for a Markov decision process (MDP) in which the reward function is learned from pairwise queries of a domain expert.
•Posted Content
Learning an Urban Air Mobility Encounter Model from Expert Preferences
TL;DR: A method for generating realistic encounter trajectories with only a few minutes of an expert's time is demonstrated, taking the form of a stochastic policy for a Markov decision process (MDP) in which the reward function is learned from pairwise queries of a domain expert.
Adaptive Coverage Path Planning for Efficient Exploration of Unknown Environments
Amanda Bouman,Joshua Ott,Sung-Hyun Kim,Kenny Chen,Mykel J. Kochenderfer,Brett T. Lopez,Ali-akbar Agha-mohammadi,Joel W. Burdick +7 more
- 23 Oct 2022
TL;DR: In this paper , a tree-based sequential decision-making process is proposed to solve the coverage problem with the objective of autonomously exploring an unknown environment under mission time constraints, where the robot is tasked with planning a path over a horizon such that the accumulated area swept out by its sensor footprint is maximized.
SAVME: Efficient Safety Validation for Autonomous Systems Using Meta-Learning
Marc R. Schlichting,Nina V. Board,Anthony Joseph Corso,Mykel J. Kochenderfer +3 more
TL;DR: A Bayesian approach that integrates meta-learning strategies with a multi-armed bandit framework is proposed, which achieves a significant speedup, up to 18 times faster compared to traditional methods that solely rely on a high-fidelity simulator.
•Proceedings Article
Partially-controlled markov decision processes for collision avoidance systems
Mykel J. Kochenderfer,James P. Chryssanthacopoulos +1 more
- 21 Aug 2018
TL;DR: This paper presents an approach that can greatly reduce the complexity of computing the optimal strategy in problems where only some of the dimensions of the problem are controllable.