CVOct 27, 2021

Feature and Label Embedding Spaces Matter in Addressing Image Classifier Bias

arXiv:2110.14336v116 citations
Originality Incremental advance
AI Analysis

This addresses bias in image classifiers for applications like facial recognition, but it is incremental as it builds on existing methods for bias mitigation.

The paper tackled image classifier bias by identifying a bias direction in the feature space and mitigating it through label embedding spaces, showing effectiveness in improving fairness while preserving classification performance on biased datasets.

This paper strives to address image classifier bias, with a focus on both feature and label embedding spaces. Previous works have shown that spurious correlations from protected attributes, such as age, gender, or skin tone, can cause adverse decisions. To balance potential harms, there is a growing need to identify and mitigate image classifier bias. First, we identify in the feature space a bias direction. We compute class prototypes of each protected attribute value for every class, and reveal an existing subspace that captures the maximum variance of the bias. Second, we mitigate biases by mapping image inputs to label embedding spaces. Each value of the protected attribute has its projection head where classes are embedded through a latent vector representation rather than a common one-hot encoding. Once trained, we further reduce in the feature space the bias effect by removing its direction. Evaluation on biased image datasets, for multi-class, multi-label and binary classifications, shows the effectiveness of tackling both feature and label embedding spaces in improving the fairness of the classifier predictions, while preserving classification performance.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes