PRISM: A Rich Class of Parameterized Submodular Information Measures for Guided Subset Selection

TL;DR

PRISM introduces a versatile class of parameterized submodular information measures that enable guided subset selection, balancing diversity, representation, and similarity to meet specific data-driven objectives.

Contribution

It proposes a novel, flexible framework for guided subset selection using parameterized submodular functions, generalizing previous methods and demonstrating effectiveness in real-world tasks.

Findings

PRISM outperforms state-of-the-art methods in targeted learning tasks.

PRISM achieves superior results in guided image-collection summarization.

The framework offers a broad modeling capability for diverse subset selection applications.

Abstract

With ever-increasing dataset sizes, subset selection techniques are becoming increasingly important for a plethora of tasks. It is often necessary to guide the subset selection to achieve certain desiderata, which includes focusing or targeting certain data points, while avoiding others. Examples of such problems include: i)targeted learning, where the goal is to find subsets with rare classes or rare attributes on which the model is underperforming, and ii)guided summarization, where data (e.g., image collection, text, document or video) is summarized for quicker human consumption with specific additional user intent. Motivated by such applications, we present PRISM, a rich class of PaRameterIzed Submodular information Measures. Through novel functions and their parameterizations, PRISM offers a variety of modeling capabilities that enable a trade-off between desired qualities of a subset like diversity or representation and similarity/dissimilarity with a set of data points. We demonstrate how PRISM can be applied to the two real-world problems mentioned above, which require guided subset selection. In doing so, we show that PRISM interestingly generalizes some past work, therein reinforcing its broad utility. Through extensive experiments on diverse datasets, we demonstrate the superiority of PRISM over the state-of-the-art in targeted learning and in guided image-collection summarization