# ELSA: Efficient Label Shift Adaptation through the Lens of Semiparametric Models

arXiv: https://arxiv.org/abs/2305.19123 · HTML: https://mlsanity.com/paper/2305.19123/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 20 references that are in the map.
- First posted on arXiv 30 May 2023.

## Builds on (most cited first, 10 of 10)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Detecting and Correcting for Label Shift with Black Box Predictors](https://mlsanity.com/paper/1802.03916/) (2018)
- [A review of domain adaptation without target labels](https://mlsanity.com/paper/1901.05335/) (2019)
- [Regularized Learning for Domain Adaptation under Label Shifts](https://mlsanity.com/paper/1903.09734/) (2019)
- [A Unified View of Label Shift Estimation](https://mlsanity.com/paper/2003.07554/) (2020)
- [Domain Adaptation under Open Set Label Shift](https://mlsanity.com/paper/2207.13048/) (2022)
- [Minimax optimal approaches to the label shift problem in non-parametric settings](https://mlsanity.com/paper/2003.10443/) (2020)
- [Unsupervised Learning under Latent Label Shift](https://mlsanity.com/paper/2207.13179/) (2022)
- [Inference with generalizable classifier predictions](https://mlsanity.com/paper/2106.07623/) (2021)

## Built on by (most cited first, 9 of 9)

- [Positive and Unlabeled Data: Model, Estimation, Inference, and Classification](https://mlsanity.com/paper/2407.09735/) (2024)
- [Estimating calibration error under label shift without labels](https://mlsanity.com/paper/2312.08586/) (2023)
- [ReTaSA: A Nonparametric Functional Estimation Approach for Addressing Continuous Target Shift](https://mlsanity.com/paper/2401.16410/) (2024)
- [Open Set Label Shift with Test Time Out-of-Distribution Reference](https://mlsanity.com/paper/2505.05868/) (2025)
- [Distribution-Free Fair Federated Learning with Small Samples](https://mlsanity.com/paper/2402.16158/) (2024)
- [Comments on Friedman's Method for Class Distribution Estimation](https://mlsanity.com/paper/2405.16666/) (2024)
- [DNA-SE: Towards Deep Neural-Nets Assisted Semiparametric Estimation](https://mlsanity.com/paper/2408.02045/) (2024)
- [Unsupervised Domain Adaptation for Binary Classification with an Unobservable Source Subpopulation](https://mlsanity.com/paper/2509.20587/) (2025)
- [From matrix inversion to constraints: provably tighter confidence regions for importance weights in label shift](https://mlsanity.com/paper/2609.14802/) (2026)

## Abstract (arXiv)

We study the domain adaptation problem with label shift in this work. Under the label shift context, the marginal distribution of the label varies across the training and testing datasets, while the conditional distribution of features given the label is the same. Traditional label shift adaptation methods either suffer from large estimation errors or require cumbersome post-prediction calibrations. To address these issues, we first propose a moment-matching framework for adapting the label shift based on the geometry of the influence function. Under such a framework, we propose a novel method named underlineEfficient underlineLabel underlineShift underlineAdaptation (ELSA), in which the adaptation weights can be estimated by solving linear systems. Theoretically, the ELSA estimator is √n-consistent (n is the sample size of the source data) and asymptotically normal. Empirically, we show that ELSA can achieve state-of-the-art estimation performances without post-prediction calibrations, thus, gaining computational efficiency.