Artificial Intelligence Colloquium: Robots Learning Through Interaction

Thursday, May 25, 2023, 4:00pm

Robots Learning Through Interaction

• Jens Kober - TU Delft
• Place: SuperC/Generali-Saal

 

Abstract

The acquisition and self-improvement of novel motor skills is among the most important problems in robotics. Complexity arises from interactions with their environment and humans, dealing with high-dimensional input data, non-linear dynamics in general and contacts in particular, multiple reference frames, and variability in objects, environments, tasks, and human behavior. A human teacher is always involved in the learning process, either directly (providing data) or indirectly (designing the optimization criterion), which raises the question: How to best make use of the interactions with the human teacher to render the learning process efficient and effective? In this talk I’ll argue that there are tremendous benefits in having a human teacher intermittently interact with a robot also while it is learning. I will discuss various methods we have developed in the fields of supervised learning, imitation learning, reinforcement learning, and interactive learning. All these concepts will be illustrated with benchmark tasks and real robot experiments ranging from fun (ball-in-a-cup) to more applied (retail environments).

The talk will be held in person but also live streamed via Zoom. The event is free, but a registration is required, and the link can be found on our web page.

After the talk, there will be a networking session in the foyer with drinks and nibbles and all are welcome to stay.