Fair and Diverse DPP-based Data Summarization

Abstract

Sampling methods that choose a subset of the data proportional to its diversity in the feature space are popular for data summarization. However, recent studies have noted the occurrence of bias (under- or over-representation of a certain gender or race) in such data summarization methods. In this paper we initiate a study of the problem of outputting a diverse and fair summary of a given dataset. We work with a well-studied determinantal measure of diversity and corresponding distributions (DPPs) and present a framework that allows us to incorporate a general class of fairness constraints into such distributions. Coming up with efficient algorithms to sample from these constrained determinantal distributions, however, suffers from a complexity barrier and we present a fast sampler that is provably good when the input vectors satisfy a natural property. Our experimental results on a real-world and an image dataset show that the diversity of the samples produced by adding fairness constraints is not too far from the unconstrained case, and we also provide a theoretical explanation of it.

arXiv commentA short version of this paper appeared in the workshop FAT/ML 2016 - arXiv:1610.07183

Cited by in corpus (16)

• Classification with Fairness Constraints: A Meta-Algorithm with Provable Guarantees
• Finding Fair and Efficient Allocations When Valuations Don't Add Up
• Maximum Entropy Distributions: Bit Complexity and Stability
• Fairness in Streaming Submodular Maximization: Algorithms and Hardness
• Equalized odds postprocessing under imperfect group information
• FAIR: Fairness-Aware Information Retrieval Evaluation
• Streaming Algorithms for Diversity Maximization with Fairness Constraints
• Diverse Data Selection under Fairness Constraints
• Unbiased estimators for random design regression
• Quality-Diversity Generative Sampling for Learning with Synthetic Data
• On the Parameterized Intractability of Determinant Maximization
• Max-Min Diversification with Fairness Constraints: Exact and Approximation Algorithms
• Spanning Tree Constrained Determinantal Point Processes are Hard to (Approximately) Evaluate
• Flexible Modeling of Diversity with Strongly Log-Concave Distributions
• Pairwise Fairness for Ordinal Regression
• Computational Complexity of Normalizing Constants for the Product of Determinantal Point Processes