ICML 2018oral149 citations

Fair and Diverse DPP-Based Data Summarization

Elisa Celis, Vijay Keswani, Damian Straszak, Amit Deshpande, Tarun Kathuria, Nisheeth Vishnoi

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 {–} e.g., under or over representation of a particular gender or ethnicity {–} 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. Designing efficient algorithms to sample from these constrained determinantal distributions, however, suffers from a complexity barrier; we present a fast sampler that is provably good when the input vectors satisfy a natural property. Our empirical results on both real-world and synthetic datasets show that the diversity of the samples produced by adding fairness constraints is not too far from the unconstrained case.

BibTeX
@InProceedings{pmlr-v80-celis18a,
  title = 	 {Fair and Diverse {DPP}-Based Data Summarization},
  author =       {Celis, Elisa and Keswani, Vijay and Straszak, Damian and Deshpande, Amit and Kathuria, Tarun and Vishnoi, Nisheeth},
  booktitle = 	 {Proceedings of the 35th International Conference on Machine Learning},
  pages = 	 {716--725},
  year = 	 {2018},
  editor = 	 {Dy, Jennifer and Krause, Andreas},
  volume = 	 {80},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {10--15 Jul},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v80/celis18a/celis18a.pdf},
  url = 	 {https://proceedings.mlr.press/v80/celis18a.html},
  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 {–} e.g., under or over representation of a particular gender or ethnicity {–} 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. Designing efficient algorithms to sample from these constrained determinantal distributions, however, suffers from a complexity barrier; we present a fast sampler that is provably good when the input vectors satisfy a natural property. Our empirical results on both real-world and synthetic datasets show that the diversity of the samples produced by adding fairness constraints is not too far from the unconstrained case.}
}
Fair and Diverse DPP-Based Data Summarization · ICML 2018