ECE Seminar Speaker - Eshed Ohn-Bar

Machines That Learn by Interacting with Humans

Eshed Ohn-Bar Humboldt Research Fellow at the Max Planck Institute for Intelligent Systems

Faculty Host:Michel Kinsy

Refreshments at 10:45am

Abstract: How can we design data-driven, computational algorithms that effectively reason over the behavior of humans in real-world, safety-critical, shared autonomy scenarios? Towards answering this fundamental question, I will show how to achieve effective and integrated system perception and action in embodied, human-aware systems, with applications to cyber-physical systems in transportation, healthcare, and accessibility. First, focusing on human-centered machine perception, I will propose a human-like, attention-based perception framework that enables learning functional and contextual models. The proposed framework leverages human guidance during training to learn a notion of situational awareness, while also mitigating dataset bias and generalization issues. I will demonstrate the learned perception models to be particularly suitable for safety-critical tasks, such as mobility and human interaction. Second, focusing on machine action, I will present a data-efficient interactive learning framework for assistive systems, and study it in the context of assistive navigation of a blind person. The general learning framework is specifically suited for collaboration with diverse users and environments. Based on real-world experimental analysis, I will show how the proposed learning framework efficiently adapts to diverse blind users while enabling long-term prediction and planning over user behavior. Together, the two frameworks provide a step towards perception-action cyber-physical systems that can learn from and operate around humans.

Bio: Eshed Ohn-Bar is a Humboldt research fellow at the Max Planck Institute for Intelligent Systems. Previously, he was a post-doc at the Computer Vision Group and Cognitive Assistance Lab in the Robotics Institute at CMU. His work has been awarded the 2017 best PhD dissertation award from the IEEE Intelligent Transportation Systems Society, a double best student paper award honorable mention at ICPR 2016, the best industry related paper award honorable mention at ICPR 2014, and the best paper award at the IEEE Workshop on Analysis and Modeling of Faces and Gestures at CVPR 2013. He co-organized several workshops on machine perception and learning for intelligent vehicles at CVPR, ICCV, and IV conferences, and is an Associate Editor for the IEEE Intelligent Vehicles Symposium 2019. Eshed received the BS degree in mathematics from UCLA in 2010, MEd from UCLA in 2011, and the PhD degree in electrical engineering fromUCSD in 2017.