Fairness Through Controlled (Un)Awareness in Node Embeddings

Graph representation learning is central for the application of machine learning (ML) models to complex graphs, such as social networks. Ensuring `fair' representations is essential, due to the societal implications and the use of sensitive personal data. In this paper, we demonstrate how the parametrization of the \emph{CrossWalk} algorithm influences the ability to infer a sensitive attributes from node embeddings. By fine-tuning hyperparameters, we show that it is possible to either significantly enhance or obscure the detectability of these attributes. This functionality offers a valuable tool for improving the fairness of ML systems utilizing graph embeddings, making them adaptable to different fairness paradigms.

• Publication year

  2024

• Venue

  arXiv.org

• Publication date

  2024-07-29

• Fields of study

  Computer Science

• Identifiers
  DOI 10.48550/arXiv.2407.20024arXiv 2407.20024
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Measurement and analysis of implied identity in ad delivery optimization

  2022cited by this paper
• CrossWalk: Fairness-enhanced Node Representation Learning

  2021cited by this paper
• Distributive Justice and Fairness Metrics in Automated Decision-making: How Much Overlap Is There?

  2021cited by this paper
• Understanding graph embedding methods and their applications

  2020cited by this paper
• A Survey on Bias and Fairness in Machine Learning

  2019cited by this paper
• Counterfactual Fairness

  2017cited by this paper
• node2vec: Scalable Feature Learning for Networks

  2016influential reference
• The Case for Process Fairness in Learning: Feature Selection for Fair Decision Making

  2016cited by this paper
• Big Data's Disparate Impact

  2016cited by this paper
• “Justice” and “fairness” are not the same thing

  2015cited by this paper
• Measuring Justice and Fairness

  2015cited by this paper
• DeepWalk: online learning of social representations

  2014influential reference
• Justice

  2013cited by this paper
• Efficient Estimation of Word Representations in Vector Space

  2013cited by this paper
• DATA ANALYSIS IN PUBLIC SOCIAL NETWORKS

  2012influential reference
• Learning from labeled and unlabeled data with label propagation

  2002cited by this paper
• A Theory of Justice

  1971cited by this paper
• Independent High-Level Expert Group on Artificial Intelligence)

  year unknowncited by this paper

• No citing papers are available for this paper.