Simultaneous Unlearning of Multiple Protected User Attributes From Variational Autoencoder Recommenders Using Adversarial Training

Gustavo Escobedo, Christian Ganhör, Stefan Brandl, Mirjam Augstein, Markus Schedl

Research output: Chapter in Book/Report/Conference proceedingsConference contributionpeer-review

Abstract

In widely used neural network-based collaborative filtering models, users’ history logs are encoded into latent embeddings that represent the users’ preferences. In this setting, the models are capable of mapping users’ protected attributes (e.g., gender or ethnicity) from these user embeddings even without explicit access to them, resulting in models that may treat specific demographic user groups unfairly and raise privacy issues. While prior work has approached the removal of a single protected attribute of a user at a time, multiple attributes might come into play in real-world scenarios. In the work at hand, we present AdvXMultVAE which aims to unlearn multiple protected attributes (exemplified by gender and age) simultaneously to improve fairness across demographic user groups. For this purpose, we couple a variational autoencoder (VAE) architecture with adversarial training (AdvMultVAE) to support simultaneous removal of the users’ protected attributes with continuous and/or categorical values. Our experiments on two datasets, LFM-2b-100k and Ml-1m, from the music and movie domains, respectively, show that our approach can yield better results than its singular removal counterparts (based on AdvMultVAE) in effectively mitigating demographic biases whilst improving the anonymity of latent embeddings.

Original languageEnglish
Title of host publicationAdvances in Bias and Fairness in Information Retrieval - 5th International Workshop, BIAS 2024, Revised Selected Papers
EditorsAlejandro Bellogin, Ludovico Boratto, Francesca Maridina Malloci, Mirko Marras, Styliani Kleanthous, Elisabeth Lex
PublisherSpringer
Pages91-102
Number of pages12
ISBN (Print)9783031719745
DOIs
Publication statusPublished - 2025
Event5th International Workshop on Algorithmic Bias in Search and Recommendation, BIAS 2024 - Washington, United States
Duration: 18 Jul 202418 Jul 2024

Publication series

NameCommunications in Computer and Information Science
Volume2227 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference5th International Workshop on Algorithmic Bias in Search and Recommendation, BIAS 2024
Country/TerritoryUnited States
CityWashington
Period18.07.202418.07.2024

Keywords

  • Collaborative Filtering
  • Debiasing
  • Privacy
  • Recommender Systems
  • Variational Autoencoder

Fingerprint

Dive into the research topics of 'Simultaneous Unlearning of Multiple Protected User Attributes From Variational Autoencoder Recommenders Using Adversarial Training'. Together they form a unique fingerprint.

Cite this