[CFAR Rising Star Lecture Series]Robust and Safe Artificial Intelligence by David Bossens
6 Jan 2023 | 5.00pm (Singapore Time)
Robustness and safety are of key importance when designing intelligent systems that learn, especially when such systems are interacting with a complex environment. In this talk, David Bossens will provide an overview of his past and ongoing research in improving the robustness and safety of AI systems.
Drawing on techniques such as neural networks, reinforcement learning, evolutionary algorithms, and Bayesian optimisation would be of interest to a wide range of machine learning enthusiasts. This talk will mainly cover robotics, navigation, and game playing as motivating application areas.
SPEAKER
David Bossens is a Research Fellow at the University of Southampton as part of the Trustworthy Autonomous Systems Hub. He received his PhD in Artificial Intelligence at the University of Southampton where he focused on reinforcement learning with limited prior knowledge in long-term environments. He was previously a Research Fellow for Resilient Robot Swarms also at the University of Southampton, where he focused on evolutionary algorithms and policy search for rapid adaptation in the context of faults and other unexpected environmental changes.
At present, his research focus is on designing reinforcement learning systems that are safe and robust. David's work has been published in leading journals and conferences, including Neural Networks, IEEE Transactions on Evolutionary Computation, GECCO, ICRA, and Machine Learning. David was also the lead organiser of the Safe Reinforcement Learning Workshop at the IJCAI 2022 conference.