# Enforcing Conditional Independence for Fair Representation Learning and Causal Image Generation

arXiv: https://arxiv.org/abs/2404.13798 · HTML: https://mlsanity.com/paper/2404.13798/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 37 references that are in the map.
- First posted on arXiv 21 Apr 2024.

## Builds on (most cited first, 22 of 22)

- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Equality of Opportunity in Supervised Learning](https://mlsanity.com/paper/1610.02413/) (2016)
- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [Counterfactual Fairness](https://mlsanity.com/paper/1703.06856/) (2017)
- [Mitigating Unwanted Biases with Adversarial Learning](https://mlsanity.com/paper/1801.07593/) (2018)
- [Diffusion Autoencoders: Toward a Meaningful and Decodable Representation](https://mlsanity.com/paper/2111.15640/) (2021)
- [Learning Adversarially Fair and Transferable Representations](https://mlsanity.com/paper/1802.06309/) (2018)
- [Learning Not to Learn: Training Deep Neural Networks with Biased Data](https://mlsanity.com/paper/1812.10352/) (2018)
- [Invariant Representations without Adversarial Training](https://mlsanity.com/paper/1805.09458/) (2018)
- [D2C: Diffusion-Denoising Models for Few-shot Conditional Generation](https://mlsanity.com/paper/2106.06819/) (2021)
- [Fully-adaptive Feature Sharing in Multi-Task Networks with Applications in Person Attribute Classification](https://mlsanity.com/paper/1611.05377/) (2016)
- [Conditional Learning of Fair Representations](https://mlsanity.com/paper/1910.07162/) (2019)
- [Adversarial Invariant Feature Learning with Accuracy Constraint for Domain Generalization](https://mlsanity.com/paper/1904.12543/) (2019)
- [Few-Shot Diffusion Models](https://mlsanity.com/paper/2205.15463/) (2022)
- [Learning Fair Representations for Kernel Models](https://mlsanity.com/paper/1906.11813/) (2019)
- [The DeCAMFounder: Non-Linear Causal Discovery in the Presence of Hidden Variables](https://mlsanity.com/paper/2102.07921/) (2021)
- [Metadata Normalization](https://mlsanity.com/paper/2104.09052/) (2021)
- [Causal Imputation via Synthetic Interventions](https://mlsanity.com/paper/2011.03127/) (2020)
- [Through the Data Management Lens: Experimental Analysis and Evaluation of Fair Classification](https://mlsanity.com/paper/2101.07361/) (2021)
- [A Penalty Approach for Normalizing Feature Distributions to Build Confounder-Free Models](https://mlsanity.com/paper/2207.04607/) (2022)
- [OTClean: Data Cleaning for Conditional Independence Violations using Optimal Transport](https://mlsanity.com/paper/2403.02372/) (2024)
- [Confounder-Aware Visualization of ConvNets](https://mlsanity.com/paper/1907.12727/) (2019)

## Built on by (most cited first, 4 of 4)

- [CAD-VAE: Leveraging Correlation-Aware Latents for Comprehensive Fair Disentanglement](https://mlsanity.com/paper/2503.07938/) (2025)
- [Diffusion Counterfactual Generation with Semantic Abduction](https://mlsanity.com/paper/2506.07883/) (2025)
- [Revisiting Diffusion Autoencoder Training for Image Reconstruction Quality](https://mlsanity.com/paper/2504.21368/) (2025)
- [MIFair: A Mutual-Information Framework for Intersectionality and Multiclass Fairness](https://mlsanity.com/paper/2604.28030/) (2026)

## Abstract (arXiv)

Conditional independence (CI) constraints are critical for defining and evaluating fairness in machine learning, as well as for learning unconfounded or causal representations. Traditional methods for ensuring fairness either blindly learn invariant features with respect to a protected variable (e.g., race when classifying sex from face images) or enforce CI relative to the protected attribute only on the model output (e.g., the sex label). Neither of these methods are effective in enforcing CI in high-dimensional feature spaces. In this paper, we focus on a nascent approach characterizing the CI constraint in terms of two Jensen-Shannon divergence terms, and we extend it to high-dimensional feature spaces using a novel dynamic sampling strategy. In doing so, we introduce a new training paradigm that can be applied to any encoder architecture. We are able to enforce conditional independence of the diffusion autoencoder latent representation with respect to any protected attribute under the equalized odds constraint and show that this approach enables causal image generation with controllable latent spaces. Our experimental results demonstrate that our approach can achieve high accuracy on downstream tasks while upholding equality of odds.