Melih Kandemir (University of Southern Denmark): Embodied Estimators for Full Autonomy

Time and place:
February 20th, 14:15
Lecture room T5, CS Buidling, Aalto University (in person) and Zoom

Abstract: A fully autonomous agent is a system that is capable of both changing its environment and adapting to it during its whole life time. Despite all the breakthroughs we saw in AI technologies, we are still missing fully autonomous agents. The stellar AI artifacts surrounding us such as robotic systems, language models, pattern recognizers, and deepfake generators are at present being trained in hard isolation from their target environments. I will attempt to characterize the barriers the actual machine learning paradigm sets in front of the development of fully autonomous systems using reinforcement learning concepts. I will present my key hypothesis that we need embodied estimators to remove these barriers. I will introduce a few baby steps my lab has recently taken to develop such estimators and conclude with some food for thought about future research.

Speaker:  Melih Kandemir is an associate professor at University of Southern Denmark (SDU). He is the founding PI of the SDU Adaptive Intelligence Laboratory (ADIN Lab), which practices fundamental research on probabilistic approaches to reinforcement learning with specific focus on continuous adaptive control applications. Melih is also the founding head of the SDU Centre for Machine Learning. Prior to his current role, Melih was leading a research group at Bosch Center for Artificial Intelligence in Renningen, Germany. Melih's current research is funded by the Novo Nordisk Foundation, Carlsberg Foundation, and the Independent Research Fund Denmark.