Disentangled Representation with Causal Constraints for Counterfactual Fairness
This addresses fairness in machine learning for real-world applications by moving beyond correlation-based methods, though it is incremental in integrating causal constraints into existing frameworks.
The paper tackled the problem of learning fair representations by incorporating causal relationships between latent variables to achieve counterfactual fairness, proposing CF-VAE which showed improved fairness and accuracy compared to benchmarks.
Much research has been devoted to the problem of learning fair representations; however, they do not explicitly the relationship between latent representations. In many real-world applications, there may be causal relationships between latent representations. Furthermore, most fair representation learning methods focus on group-level fairness and are based on correlations, ignoring the causal relationships underlying the data. In this work, we theoretically demonstrate that using the structured representations enable downstream predictive models to achieve counterfactual fairness, and then we propose the Counterfactual Fairness Variational AutoEncoder (CF-VAE) to obtain structured representations with respect to domain knowledge. The experimental results show that the proposed method achieves better fairness and accuracy performance than the benchmark fairness methods.