Reducing Unintended Bias of ML Models on Tabular and Textual Data - Archive ouverte HAL Access content directly
Conference Papers Year : 2021

Reducing Unintended Bias of ML Models on Tabular and Textual Data

Guilherme Alves
Maxime Amblard
Fabien Bernier
  • Function : Author
  • PersonId : 1083723
Miguel Couceiro
Amedeo Napoli

Abstract

Unintended biases in machine learning (ML) models are among the major concerns that must be addressed to maintain public trust in ML. In this paper, we address process fairness of ML models that consists in reducing the dependence of models' on sensitive features, without compromising their performance. We revisit the framework FIXOUT that is inspired in the approach "fairness through unawareness" to build fairer models. We introduce several improvements such as automating the choice of FIXOUT's parameters. Also, FIXOUT was originally proposed to improve fairness of ML models on tabular data. We also demonstrate the feasibility of FIXOUT's workflow for models on textual data. We present several experimental results that illustrate the fact that FIXOUT improves process fairness on different classification settings.
Fichier principal
Vignette du fichier
_dsaa_21__Reducing_Unintended_Bias_of_ML_Models_on_Tabular_and_Textual_Data (5).pdf (340.14 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03312797 , version 1 (02-08-2021)
hal-03312797 , version 2 (06-09-2021)

Identifiers

  • HAL Id : hal-03312797 , version 2

Cite

Guilherme Alves, Maxime Amblard, Fabien Bernier, Miguel Couceiro, Amedeo Napoli. Reducing Unintended Bias of ML Models on Tabular and Textual Data. DSAA 2021 - 8th IEEE International Conference on Data Science and Advanced Analytics, Oct 2021, Porto (virtual event), Portugal. pp.1-10. ⟨hal-03312797v2⟩
174 View
228 Download

Share

Gmail Facebook Twitter LinkedIn More