Contact
Dr.-Ing. Jens Kober
Associate Professor, Cognitive Robotics department (CoR)
Delft University of Technology (TU Delft)
Mekelweg 2 (3mE Building)
2628 CD Delft, The Netherlands
🖂 j.kober (at) tudelft.nl
🕿 +31 (0)15 27 85150
Office: F-2-380
https://www.linkedin.com/in/jens-kober/
Research interests
Robotics, Motor Skill Learning, Machine Learning, Reinforcement Learning, Imitation Learning, Deep Learning, Interactive Learning
News
- new preprint:
. Interactive Imitation Learning in State-Space. arXiv:2008.00524 [cs.RO], 2020.
- new journal paper:
. Learning Sequential Force Interaction Skills. Robotics, 9(2):45, 2020.
- new report about our NWO OpenMind project (Dutch only) Hoe trek je een kies?
- new report about our NWO OpenMind project Understanding tooth extraction thanks to robot
- new conference paper:
. Robot Technology in Analyzing Tooth Removal – a Proof of Concept. In Annual International Conferences of the IEEE Engineering in Medicine and Biology Society (EMBC), 2020.
- new journal paper:
. Interactive Learning of Temporal Features for Control: Shaping Policies and State Representations From Human Feedback. IEEE Robotics & Automation Magazine, 27(2):46–54, 2020.
- new conference paper:
. Fine-tuning Deep RL with Gradient-Free Optimization. In IFAC World Congress, 2020.
- new conference paper:
. Predicting and Optimizing Ergonomics in Physical Human-Robot Cooperation Tasks. In IEEE International Conference on Robotics and Automation (ICRA), 2020.
- Tim de Bruin successfully defended his PhD thesis. Congrats!
- Jens Kober is co-organizing the Workshop on Interactive Robot Learning@ICRA2020
- new book chapter:
. Encyclopedia of Systems and Control (John Baillieul, Tariq Samad, eds.), chapter Robot Learning, Springer London, pp. 1–9, 2019.
- New PostDoc position “Interactive Reinforcement Learning” (30 months)
- new conference paper:
. Deep Reinforcement Learning with Feedback-based Exploration. In IEEE Conference on Decision and Control (CDC), pp. 803–808, 2019.
and corresponding blog post - new journal paper:
. Reinforcement Learning of Motor Skills using Policy Search and Human Corrective Advice. International Journal of Robotics Research, 38(14):1560–1580, 2019.
- Jens Kober will give a keynote at the World Robot Conference
- Home of Innovation article (with a fun comic): Deep Learning for Robust robot Control
- Jens Kober is co-organizing an IEEE RAM Special Issue on “Deep Learning and Machine Learning in Robotics”
- new conference paper:
. Simultaneous Learning of Objective Function and Policy from Interactive Teaching with Corrective Feedback. In IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM), pp. 726–732, 2019.
- Jens Kober will be a speaker at the RE•WORK Deep Learning Summit London
- Jens Kober has been promoted to associate professor (tenured)
- new preprints:
. Learning Gaussian Policies from Corrective Human Feedback. arXiv:1903.05216 [cs.LG], 2019.
- Jens Kober will serve as co-chair for Inclusion@RSS2019
- new conference paper:
. A Practical Bayesian Optimization Approach for the Optimal Estimation of the Rotor Effective Wind Speed. In American Control Conference (ACC), pp. 4179–4185, 2019.
- new conference paper:
. Continuous Control for High-Dimensional State Spaces: An Interactive Learning Approach. In IEEE International Conference on Robotics and Automation (ICRA), pp. 7611–7617, 2019.
- new journal paper:
. Reinforcement Learning Based Compensation Methods for Robot Manipulators. Engineering Applications of Artificial Intelligence, 78:236–247, 2019.
- new project Ahold Delhaize: AI for Retail (AIR)Lab Delft
- new project NWO OpenMind: Robot-Assistance in Understanding and Educating Tooth Removal Procedures
- new project NWO-TTW Perspective: FlexCRAFT - Cognitive Robots for Flexible Agro Food Technology
- Jens Kober will give a keynote at the 2018 Great Bay Area Conference on Robotics and Artificial Intelligence
- Jens Kober has been elevated to IEEE Senior Member
- new conference paper:
. Head-tracked off-axis perspective projection improves gaze readability of 3D virtual avatars. In SIGGRAPH Asia Technical Briefs, pp. 29:1–29:4, 2018.
- new journal paper:
. Reinforcement Learning for Control: Performance, Stability, and Deep Approximators. Annual Reviews in Control, 46:8–28, 2018.
- new conference paper:
. 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.
- Jens Kober was awarded an ERC Starting Grant
- new journal paper:
. Experience Selection in Deep Reinforcement Learning for Control. Journal of Machine Learning Research, 19(9):1–56, 2018.
- new journal paper:
- new conference paper:
. Human-Robot Cooperative Object Manipulation with Contact Changes. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 1354–1360, 2018.
- Jens Kober received the IEEE-RAS Early Academic Career Award in Robotics and Automation 2018
- new workshop paper:
. Sample-efficient Reinforcement Learning via Difference Models. In Third Machine Learning in Planning and Control of Robot Motion Workshop at IEEE International Conference on Robotics and Automation (ICRA), 2018.
- TU Delft communications: An old robot operating system learning new tricks
- Jens Kober is co-organizing the DISC Summer School on “Machine Learning for Control”
- 2 new journal papers:
. Mixture of Attractors: A Novel Movement Primitive Representation for Learning Motor Skills From Demonstrations. IEEE Robotics and Automation Letters, 3(2):926–933, 2018.
. Integrating State Representation Learning into Deep Reinforcement Learning. IEEE Robotics and Automation Letters, 3(3):1394–1401, 2018. The contents of this paper were also selected by ICRA'18 Program Committee for presentation at the Conference.
- new ICRA workshop:
Full-day Workshop on Ergonomic Physical Human-Robot Collaboration - new workshop paper:
. Human Corrective Advice in the Policy Search Loop. In Workshop Human-in-the-loop Robotic Manipulation: On the Influence of the Human Role, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017.
- new poster:
- new journal paper:
. Robottechnologie, is er een toekomst voor in de tandheelkunde? Quality Practice Tandheelkunde 12(5):30–35, 2017.
- Jens Kober is invited speaker at Philips' Strategy Day, Eindhoven, Netherlands
- Jens Kober is invited speaker at 2nd Workshop on Machine Learning Control, LAMIH, Université de Valenciennes, France
- new workshop paper:
. Off Policy Experience Retention for Deep Actor Critic Learning. In Deep Reinforcement Learning Workshop, Advances in Neural Information Processing Systems (NIPS), 2016.
- new abstract:
. Op zoek naar toepassingen van robottechnologie in de MKA-chirurgie. In Annual Scientific Meeting of the Dutch Association of Oral and Maxillofacial Surgery, 2016.
- Jens Kober is invited speaker at Global Robot Expo 2017, Madrid, Spain
- new CDC paper:
. Learning State Representation for Deep Actor-Critic Control. In IEEE Conference on Decision and Control (CDC), pp. 4667–4673, 2016.
- Jens Kober quoted in MIT Technology Review which was picked up by Forbes and The Robot Report
- 2 new IROS papers:
. Probabilistic Decomposition of Sequential Force Interaction Tasks into Movement Primitives. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 3920–3927, 2016.
. Improved Deep Reinforcement Learning for Robotics Through Distribution-based Experience Retention. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 3947–3952, 2016.
- Tim de Bruin won a Qualcomm Innovation Fellowship (QInF) Europe
- Jens Kober is invited speaker at RoboBusiness Europe 2016, Odense, Denmark
- Jens Kober is co-organizing the DISC Summer School on “A Systems and Control Perspective in Human Robot Environment Interaction”
- Jens Kober is hosting a Bèta Balie Debate, Delft, Netherlands
- Jens Kober is co-chair of the IEEE-RAS TC Robot Learning
- new PSYCO paper:
- 2 cohesie projects (interdepartmental projects within 3mE / TU Delft) granted:
“Learning-based motion control with ROS/MoveIt” Gijs van der Hoorn (BMechE), Jens Kober (DCSC)
“Artificial Muscles: a Physiological Mechatronics Approach” Cornelis van de Kamp & Jens Kober (DCSC), Heike Vallery (BMechE), Hassan HosseinNia (PME)