file.pdf (567.71 kB)
0/0

Learning Policies for Contextual Submodular Prediction

Download (567.71 kB)
journal contribution
posted on 01.05.2013 by Stephane Ross, Jiaji Zhou, Yisong Yue, Debadeepta Dey, J. Andrew Bagnell

Many prediction domains, such as ad placement, recommendation, trajectory prediction, and document summarization, require predicting a set or list of options. Such lists are often evaluated using submodular reward functions that measure both quality and diversity. We propose a simple, efficient, and provably near-optimal approach to optimizing such prediction problems based on no-regret learning. Our method leverages a surprising result from online submodular optimization: a single no-regret online learner can compete with an optimal sequence of predictions. Compared to previous work, which either learn a sequence of classifiers or rely on stronger assumptions such as realizability, we ensure both data-efficiency as well as performance guarantees in the fully agnostic setting. Experiments validate the efficiency and applicability of the approach on a wide range of problems including manipulator trajectory optimization, news recommendation and document summarization.

History

Publisher Statement

Copyright 2013 by the author(s)

Date

01/05/2013

Exports

Keyword(s)

Exports