On the Fairness of Disentangled Representations

By Francesco Locatello, Gabriele Abbati, Tom Rainforth, Stefan Bauer, Bernhard Schölkopf, Olivier Bachem. 2019

250 citations recorded by the source corpus.

Recently there has been a significant interest in learning disentangled representations, as they promise increased interpretability, generalization to unseen scenarios and faster learning on downstream tasks. In this paper, we investigate the usefulness of different notions of disentanglement for improving the fairness of downstream prediction tasks based on representations. We consider the setting where the goal is to predict a target variable based on the learned representation of high-dimensional observations (such as images) that depend on both the target variable and an \emph{unobserved} sensitive variable. We show that in this setting both the optimal and empirical predictions can be unfair, even if the target variable and the sensitive variable are independent. Analyzing the representations of more than \num{12600} trained state-of-the-art disentangled models, we observe that several disentanglement scores are consistently correlated with increased fairness, suggesting that disentanglement may be a useful property to encourage fairness when sensitive variables are not observed.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

A Sober Look at the Unsupervised Learning of Disentangled Representations and their Evaluation (2020, background): “Notable examples are applications in fairness (Locatello et al., 2019) and abstract visual reasoning (van Steenkiste et al.”
Toward Causal Representation Learning (2021, background): “Machine learning models that incorporate or learn structural knowledge of an environment have been shown to be more efﬁcient and generalize better [9], [11], [15], [16], [27], [58], [77], [84], [85], [141], [157], [177], [181], [197], [211], [212], [244], [258], [272], [274].”
Towards Causal Representation Learning (2021, background): “Much of the existing work on disentanglement [109, 158, 159, 256, 157, 135, 202, 61] focuses on independent factors”
On the Transfer of Disentangled Representations in Realistic Settings (2020, methodology): “In particular, using the architecture from Locatello et al. (2019b), none of the models we trained correctly captured all factors of variation or yielded high-quality reconstructions.”
Weakly-Supervised Disentanglement Without Compromises (2020, methodology): “Further, we show that these representations are useful on a diverse suite of downstream tasks, including a novel experiment targeting strong generalization under covariate shifts, fairness (Locatello et al., 2019a) and abstract visual reasoning (van Steenkiste et al., 2019).”
On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Dataset (2019, background): “Despite the growing importance of the ﬁeld and the potential societal impact in the medical domain or fair decision making [e.g. 6, 10, 37], the performance of state-of-the-art disentanglement learning on real-world data is unknown.”
Are Disentangled Representations Helpful for Abstract Visual Reasoning? (2019, background): “Following the successes in learning distributed representations that efficiently encode the content of high-dimensional sensory data [45, 56, 76], recent work has focused on learning representations that are disentangled [6, 69, 68, 73, 71, 26, 27, 42, 10, 63, 16, 52, 53, 48, 9, 51].”
Unsupervised Model Selection for Variational Disentangled Representation Learning (2019, methodology): “Finally, we validate that the model ranking produced by our approach correlates well with the ﬁnal task performance on two recently reported tasks: a classiﬁ-cation fairness task (Locatello et al., 2019) and a model-based reinforcement learning (RL) task (Watters et al., 2019).”
Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding (2020, background): “A disentangled representation would be valuable for a wide variety of machine learning applications, including sample efﬁciency for downstream tasks [8, 26], fairness [27, 28] and interpretability [1, 29, 30].”
Is Independence all you need? On the Generalization of Representations Learned from Correlated Data (2020, methodology): “For this we use a variant of demographic parity Dwork et al. (2012) that computes pairwise mutual information between latents and FoV (Locatello et al., 2019a).”
Structural Autoencoders Improve Representations for Generation and Transfer (2020, methodology): “…compare the performance of each of the methods using a variety of different metrics to elucidate the quality of generated samples (using our "hybrid" sampling method as well as the standard "prior" sampling): disentanglement [29, 30], fairness [31], and abstract visual reasoning capabilities [32].”
RELATE: Physically Plausible Multi-Object Scene Synthesis Using Structured Latent Spaces (2020, methodology): “[26] Francesco Locatello, Gabriele Abbati, Thomas Rainforth, Stefan Bauer, Bernhard Schölkopf, and Olivier Bachem.”
Algorithm fairness in artificial intelligence for medicine and healthcare (2023, background): “The evaluation of unsupervised VAE-based disentangled models has shown that disentanglement scores correlate with fairness metrics, bench-marked on numerous fairness-classification tasks without the need for protected attribute information 294 .”
CheXclusion: Fairness gaps in deep chest X-ray classifiers (2020, methodology): “Others have used reinforcement learning methods to control the model disparity level [9, 41], adversarial learning to generate a debiased input data [40, 47, 51], or target constructing a fair latent representation [1, 24, 26, 50].”
A Geometric Solution to Fair Representations (2020, background): “, penalizing unfair inferences [2, 11], or creating representations that do not strongly depend on protected features [14, 20, 23].”
Disentangling Influence: Using Disentangled Representations to Audit Model Predictions (2019, background): “Disentanglement has recently been shown to be a useful property for learning and evaluating fair machine learning models [5, 17].”
Say No to the Discrimination: Learning Fair Graph Neural Networks with Limited Sensitive Attribute Information (2020, methodology): “s which will eliminate the potential discrimination. 2.2 Fairness in Machine Learning Many works have been conducted to deal with the bias in the training data to achieve fairness in machine learning [3, 12, 20, 22, 23, 30, 56]. Based on which stage of the machine learning training process is revise”