Talk by Prof. Arun Rajkumar: "Learning to Rank from Pairwise Comparisons"

  • Date:

    13 December 2022 - 11:30

  • Speaker:

    Prof. Arun Rajkumar, Assistant Professor, IIT Madras India

     

    Arun Rajkumar is currently an Assistant Professor at the Computer Science and Engineering department of IIT Madras and leads the Machine Learning Theory lab at IITM. Prior to joining IIT Madras, he was a research scientist at the Xerox Research Center (now Conduent Labs), India for three years. He earned his Ph.D from the Indian Institute of Science. His areas of research include statistical learning theory, machine learning and sequential decision making. He has authored several papers in ML/AI that have appeared in conferences such as NeuriPS, ICML, ICLR, AAAI, UAI, COLT, KDD, ICDM, IJCAI, AAMAS etc.

  • Abstract:

    The problem of learning to rank from a set of pairwise comparisons has been studied widely in several research communities including ML, theoretical computer science, operations research, statistics, etc. In this talk, Professor Rajkumar will give an introduction to this problem and talk about several provably optimal algorithms for this problem along with some exciting structural results that involve understanding classes of tournament graphs. The pre-requisites for the talk are minimal and will be covered during the talk.