k – Online Policies and Fundamental Limits

10/15/2021
by   Samrat Mukhopadhyay, et al.
5

This paper introduces and studies the k problem – a generalization of the classic Prediction with Expert's Advice (i.e., the ) problem. Unlike the problem, where the learner chooses exactly one expert, in this problem, the learner selects a subset of k experts from a pool of N experts at each round. The reward obtained by the learner at any round depends on the rewards of the selected experts. The k problem arises in many practical settings, including online ad placements, personalized news recommendations, and paging. Our primary goal is to design an online learning policy having a small regret. In this pursuit, we propose (Sampled Hedge) - a framework for designing efficient online learning policies by leveraging statistical sampling techniques. We show that, for many related problems, improves upon the state-of-the-art bounds for regret and computational complexity. Furthermore, going beyond the notion of regret, we characterize the mistake bounds achievable by online learning policies for a class of stable loss functions. We conclude the paper by establishing a tight regret lower bound for a variant of the k problem and carrying out experiments with standard datasets.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/29/2019

Dying Experts: Efficient Algorithms with Optimal Regret Bounds

We study a variant of decision-theoretic online learning in which the se...
research
09/28/2022

Online Subset Selection using α-Core with no Augmented Regret

We consider the problem of sequential sparse subset selections in an onl...
research
01/11/2022

Learning what to remember

We consider a lifelong learning scenario in which a learner faces a neve...
research
06/15/2021

Online Learning with Uncertain Feedback Graphs

Online learning with expert advice is widely used in various machine lea...
research
10/27/2020

Online Learning with Primary and Secondary Losses

We study the problem of online learning with primary and secondary losse...
research
12/03/2020

Online learning with dynamics: A minimax perspective

We study the problem of online learning with dynamics, where a learner i...
research
02/01/2021

Impossible Tuning Made Possible: A New Expert Algorithm and Its Applications

We resolve the long-standing "impossible tuning" issue for the classic e...

Please sign up or login with your details

Forgot password? Click here to reset