【海韵讲座】2024年第52期- Equipping Neural Networks for the Open World: Addressing Explainability and Boundaries
报告题目：Equipping Neural Networks for the Open World: Addressing Explainability and Boundaries
主讲人： Prof. Nikhil R. Pal, Techno India University Calcutta, 印度国家科学院院士、印度国家工程院院士, IEEE Fellow
时间：2024年10月22日15:00-16:00
地点：翔安校区西部片区4号楼106
Abstract：
Many intelligent systems lack explainability, which creates challenges in critical areas like healthcare, defense, and law. These systems often make confident but incorrect decisions when faced with unfamiliar data outside their training domain, leading to potential risks in open-world scenarios. In this talk, we introduce methods to address these limitations in Multi-Layer Perceptrons (MLPs) by equipping them to abstain from making decisions in uncertain situations. We propose two practical approaches: an ensemble of neural networks and an additional "Don't know" class to enhance the reliability and robustness of the models. We also discuss the results and limitations of these methods, as well as possible directions for future research.
Biography:
Nikhil R. Pal is currently a Professor at Techno India University, Calcutta, and an Honorary Visiting Professor at South Asian University, New Delhi. He was previously a Professor in the Electronics and Communication Sciences Unit and the founding Head of the Center for Artificial Intelligence and Machine Learning at the Indian Statistical Institute, Calcutta.
His research interests span brain science, computational intelligence, machine learning, and data mining. He has published extensively and has served as the Editor-in-Chief of the IEEE Transactions on Fuzzy Systems. He is also on the editorial and advisory boards of several leading journals, including the International Journal of Approximate Reasoning and IEEE Transactions on Cybernetics.
Prof. Pal is the recipient of numerous awards, such as the 2015 IEEE Computational Intelligence Society (CIS) Fuzzy Systems Pioneer Award and the 2021 IEEE CIS Meritorious Service Award. He has delivered many plenary and keynote addresses at premier international conferences. He has also served as General Chair, Program Chair, and co-Program Chair of several conferences, and has been a Distinguished Lecturer of the IEEE CIS on multiple occasions. Additionally, he has held prominent roles in IEEE CIS, including Vice-President for Publications and President.
He is a Fellow of several distinguished academies, including the Indian National Science Academy (INSA), the Indian National Academy of Engineering (INAE), the Indian National Science Academy (INSA), the International Fuzzy Systems Association (IFSA), The World Academy of Sciences (TWAS), and IEEE.
邀请人：人工智能系 江敏教授