Abstract: Supervised learning, the harbinger of machine learning over the last decade, has had tremendous impact across application domains in recent years. However, the notion of a static trained machine learning model is becoming increasingly limiting, as these models are deployed in changing and evolving environments. Among a few related settings, continual learning has gained significant interest among practitioners to address this need of continually learning from new information — including new classes, tasks or domains, without losing the model’s effectiveness on past data. In this talk, we will briefly discuss lifelong (or continual) learning, and highlight the need to go beyond the vanilla setting to address real-world challenges. In particular, the talk will describe our efforts towards “organic lifelong learning”, viz, the ability of a machine learning model to continually learn over time with whatever information or data is available at a given point in time, including the much-needed ability of saying “I don’t know”. The talk will cover some of our recent research on open-world object detection (CVPR 2021), novel class discovery (ECCV 2022), and continual zero-shot learning (CVPR 2022), besides other work in this domain — and also share interesting real-world use cases of these research efforts.

Bio: Vineeth N Balasubramanian is an Associate Professor in the Department of Computer Science and Engineering at the Indian Institute of Technology, Hyderabad (IIT-H), India, and is currently a Fulbright-Nehru Visiting Faculty Fellow at Carnegie Mellon University. He was also the Founding Head of the Department of Artificial Intelligence at IIT-H from 2019-22. His research interests include deep learning, machine learning, and computer vision with a focus on explainability, continual learning and learning with limited labeled data. He served as a General Chair for ACML 2022, and serves as a Senior PC/Area Chair regularly for conferences such as CVPR, ICCV, AAAI, IJCAI and ECCV. He is a recipient of the Google Research Scholar Award (2021), NASSCOM AI Gamechanger Award (2022, both Winner and Runner-up), Teaching Excellence Award at IIT-H (2017 and 2021), Research Excellence Award at IIT-H (2022), among others. For more details, please see https://iith.ac.in/~vineethnb/.

Speaker Homepage: https://people.iith.ac.in/vineethnb/

Sponsored in part by:   Meta Reality Labs Pittsburgh