A Submodular Optimization Framework for Data, Feature and Topic Summarization

August 29, 1:45 pm - 5:00 pm (CEST)

Speakers: Rishabh Iyer and Ganesh Ramakrishnan

Tutorial website: https://sites.google.com/view/ecaitutorial2020summ/home

Agenda: A Submodular Optimization Framework for Data, Feature and Topic Summarization

This tutorial will address several applications of Data Selection and Summarization including Imageand Video Summarization, Data Subset Selection, Feature Selection, Selection problem in Graphs andRules, and Data selection in Active Learning and Meta Learning scenarios. We shall view all theseproblems from a lens of combinatorial optimization, and in particular, submodular optimizationwhich has received a lot of attention in machine learning, robotics, game theory and social networks.

We shall study a rich class of submodular functions and corresponding optimization and learning algorithms, and particularly study their utility in a) modeling aspects such as representation, coverage, and diversity for summarization problems, b) modeling the information of data and feature subsets for data subset selection, feature selection and active learning, and c) defining a class of sparsity inducing norms defined via combinatorial functions for feature and rule selection.

Rishabh Iyer: Rishabh Iyer is currently an Assistant Professor at University of Texas at Dallas. Prior to this, he was a Research Scientist at Microsoft, where he works on several problems around Online Learning, Contextual Bandits, Reinforcement Learning etc. He finished his PostDoc and Ph.D from the University of Washington, Seattle. His work has received best paper awards at ICML and NIPS, 2013. He also won the Microsoft Ph.D fellowship, a Facebook Ph.D Fellowship, and the Yang Outstanding Doctoral Student Award from University of Washington. He completed his B.Tech at the Department of from IIT Bombay in 2011, and has been a visitor at Microsoft Research, Redmond and Simon Fraser University. He has worked on several aspects of Machine Learning including Discrete and Convex optimization, deep learning, video/image summarization, data subset selection, active learning, online learning etc. He has applied his work in several domains including search advertisement, computer vision, text classification and speech.

Prof. Ganesh Ramakrishnan: Ganesh is currently serving as a Professor at the Department of Computer Science and Engineering, IIT Bombay. After completing his BTech in the year 2000 at the Department of Computer Science and Engineering IIT Bombay, he decided to remain in India and contribute. He completed his PhD also department of CSE, IIT Bombay, worked at IBM India Research labs and thereafter joined back IIT Bombay as a faculty member of the Dept of CSE in 2009. His areas of research include Human Assisted Machine Learning, symbolic representation in machine learning and computational linguistics in Indian languages. In the past, he has received awards such as IBM Faculty Award, Microsoft Research Award, Yahoo! Research Award as well as IIT Bombay Impactful Research Award. He also held the J.R. Isaac Chair at IIT Bombay. Ganesh is very passionate about boosting the AI research eco-system for India and toward that, the research by him and his students as well as collaborators has resulted in a couple of startups which he continues to mentor.