We discuss the problem of estimating the largest community (a.k.a., mode) in a population composed of multiple disjoint communities. This estimation is performed in a fixed confidence (PAC) setting via sequential sampling of individuals with replacement. We consider two sampling models: (i) an identityless model, wherein only the community of each sampled individual is revealed, and (ii) an identity-based model, wherein the learner is able to discern whether or not each sampled individual has been sampled before, in addition to the community of that individual. Our results, which are structurally similar to those available for multi-armed bandits in the PAC setting, highlight the value of identity estimation in the context of mode estimation.

This talk is based on joint work with Meera Pai and Nikhil Karamchandani.

Short Bio: Jayakrishnan Nair is an Associate Professor in the department of Electrical Engineering at IIT Bombay. His research focuses on modeling, performance evaluation, and design issues in online learning, queueing systems and communication networks, drawing on tools from stochastic modelling, queueing theory, game theory, optimization, and control theory.