University of Sussex
Browse
2008.05248.pdf (3.33 MB)

Null-sampling for interpretable and fair representations

Download (3.33 MB)
conference contribution
posted on 2023-06-09, 21:28 authored by Thomas Kehrenberg, Myles Bartlett, Oliver Thomas, Novi QuadriantoNovi Quadrianto
We propose to learn invariant representations, in the data domain, to achieve interpretability in algorithmic fairness. Invariance implies a selectivity for high level, relevant correlations w.r.t. class label annotations, and a robustness to irrelevant correlations with protected characteristics such as race or gender. We introduce a non-trivial setup in which the training set exhibits a strong bias such that class label annotations are irrelevant and spurious correlations cannot be distinguished. To address this problem, we introduce an adversarially trained model with a null-sampling procedure to produce invariant representations in the data domain. To enable disentanglement, a partially-labelled representative set is used. By placing the representations into the data domain, the changes made by the model are easily examinable by human auditors. We show the effectiveness of our method on both image and tabular datasets: Coloured MNIST, the CelebA and the Adult dataset.

Funding

BayesianGDPR - Bayesian Models and Algorithms for Fairness and Transparency; G2903; EUROPEAN UNION

History

Publication status

  • Published

File Version

  • Accepted version

Journal

Computer Vision – ECCV 2020

Publisher

Springer International Publishing

Volume

12373

Event name

European Conference on Computer Vision (ECCV)

Event location

Online

Event type

conference

Event date

23 Aug 2020 - 28 Aug 2020

ISBN

9783030586041

Series

Image Processing, Computer Vision, Pattern Recognition, and Graphics

Department affiliated with

  • Informatics Publications

Full text available

  • Yes

Peer reviewed?

  • Yes

Legacy Posted Date

2020-08-18

First Open Access (FOA) Date

2021-12-20

First Compliant Deposit (FCD) Date

2020-08-18

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC