# Class Distribution Shifts in Zero-Shot Learning: Learning Robust Representations

arXiv: https://arxiv.org/abs/2311.18575 · HTML: https://mlsanity.com/paper/2311.18575/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 61 references that are in the map.
- First posted on arXiv 30 Nov 2023.

## Builds on (most cited first, 30 of 34)

- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Matching Networks for One Shot Learning](https://mlsanity.com/paper/1606.04080/) (2016)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [Class-Balanced Loss Based on Effective Number of Samples](https://mlsanity.com/paper/1901.05555/) (2019)
- [Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss](https://mlsanity.com/paper/1906.07413/) (2019)
- [In Defense of the Triplet Loss for Person Re-Identification](https://mlsanity.com/paper/1703.07737/) (2017)
- [Out-of-Distribution Generalization via Risk Extrapolation (REx)](https://mlsanity.com/paper/2003.00688/) (2020)
- [Long-tail learning via logit adjustment](https://mlsanity.com/paper/2007.07314/) (2020)
- [Causal inference using invariant prediction: identification and confidence intervals](https://mlsanity.com/paper/1501.01332/) (2015)
- [Just Train Twice: Improving Group Robustness without Training Group Information](https://mlsanity.com/paper/2107.09044/) (2021)
- [Deep Metric Learning via Lifted Structured Feature Embedding](https://mlsanity.com/paper/1511.06452/) (2015)
- [Deep metric learning using Triplet network](https://mlsanity.com/paper/1412.6622/) (2014)
- [Balanced Meta-Softmax for Long-Tailed Visual Recognition](https://mlsanity.com/paper/2007.10740/) (2020)
- [Learning Models with Uniform Performance via Distributionally Robust Optimization](https://mlsanity.com/paper/1810.08750/) (2018)
- [A Metric Learning Reality Check](https://mlsanity.com/paper/2003.08505/) (2020)
- [Sampling Matters in Deep Embedding Learning](https://mlsanity.com/paper/1706.07567/) (2017)
- [What is the Effect of Importance Weighting in Deep Learning?](https://mlsanity.com/paper/1812.03372/) (2018)
- [Improving Calibration for Long-Tailed Recognition](https://mlsanity.com/paper/2104.00466/) (2021)
- [Statistics of Robust Optimization: A Generalized Empirical Likelihood Approach](https://mlsanity.com/paper/1610.03425/) (2016)
- [Simple data balancing achieves competitive worst-group-accuracy](https://mlsanity.com/paper/2110.14503/) (2021)
- [Mitigate Bias in Face Recognition using Skewness-Aware Reinforcement Learning](https://mlsanity.com/paper/1911.10692/) (2019)
- [On Calibration and Out-of-domain Generalization](https://mlsanity.com/paper/2102.10395/) (2021)
- [Invariant Rationalization](https://mlsanity.com/paper/2003.09772/) (2020)
- [Revisiting Training Strategies and Generalization Performance in Deep Metric Learning](https://mlsanity.com/paper/2002.08473/) (2020)
- [Context-aware Feature Generation for Zero-shot Semantic Segmentation](https://mlsanity.com/paper/2008.06893/) (2020)
- [Face Recognition: Too Bias, or Not Too Bias?](https://mlsanity.com/paper/2002.06483/) (2020)
- [Maximin effects in inhomogeneous large-scale data](https://mlsanity.com/paper/1406.0596/) (2014)
- [Signal-to-Noise Ratio: A Robust Distance Metric for Deep Metric Learning](https://mlsanity.com/paper/1904.02616/) (2019)
- [FairFace Challenge at ECCV 2020: Analyzing Bias in Face Recognition](https://mlsanity.com/paper/2009.07838/) (2020)
- [Focus on the Common Good: Group Distributional Robustness Follows](https://mlsanity.com/paper/2110.02619/) (2021)

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

- [ASAP: Unsupervised Post-training with Label Distribution Shift Adaptive Learning Rate](https://mlsanity.com/paper/2508.13445/) (2025)
- [Quantifying Uncertainty in the Presence of Distribution Shifts](https://mlsanity.com/paper/2506.18283/) (2025)
- [OASIS: Open-world Adaptive Self-supervised and Imbalanced-aware System](https://mlsanity.com/paper/2508.16656/) (2025)
- [Variational Learning of Disentangled Representations](https://mlsanity.com/paper/2506.17182/) (2025)

## Abstract (arXiv)

Zero-shot learning methods typically assume that the new, unseen classes encountered during deployment come from the same distribution as the the classes in the training set. However, real-world scenarios often involve class distribution shifts (e.g., in age or gender for person identification), posing challenges for zero-shot classifiers that rely on learned representations from training classes. In this work, we propose and analyze a model that assumes that the attribute responsible for the shift is unknown in advance. We show that in this setting, standard training may lead to non-robust representations. To mitigate this, we develop an algorithm for learning robust representations in which (a) synthetic data environments are constructed via hierarchical sampling, and (b) environment balancing penalization, inspired by out-of-distribution problems, is applied. We show that our algorithm improves generalization to diverse class distributions in both simulations and experiments on real-world datasets.