Research map / Label-efficient and robust learning / Recognition and long-tail
Null-sampling for Interpretable and Fair Representations
Thomas Kehrenberg, Myles Bartlett, Oliver Thomas, Novi Quadrianto
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use cFlow (invertible encoder) instead of cVAE as colour sigma increases | null-sampling · coloured MNIST | accuracy | Improved |
| use null-sampled representations instead of raw images for a classifier | CNN classifier · CelebA smiling, eta=0 | accuracy and DP diff | Improved |
| use null-sampled representations at intermediate bias levels | CNN classifier · CelebA smiling, intermediate mixing factors | accuracy vs DP diff | Mixed |
| use cFlow or cVAE representations instead of baselines | downstream classifier · UCI Adult, eta 0 and 1 | accuracy and fairness | Improved |
| use the Ln2L mutual-information penalty instead of grey-scaling | Ln2L baseline · coloured MNIST with strong bias | accuracy | Hurt |
Closest papers by meaning
- Invariant Representations without Adversarial Training (NeurIPS 2018 · similarity 0.89)
- Balanced Datasets Are Not Enough: Estimating and Mitigating Gender Bias in Deep Image Representations (ICCV 2019 · similarity 0.87)
- Fairness via Representation Neutralization (NeurIPS 2021 · similarity 0.87)
- Fair Representation Learning with Controllable High Confidence Guarantees via Adversarial Inference (NeurIPS 2025 · similarity 0.86)
- Classes Are Not Equal: An Empirical Study on Image Recognition Fairness (CVPR 2024 · similarity 0.85)
- Stabilizing Sample Similarity in Representation via Mitigating Random Consistency (ICML 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
