Covariate Shift

Follow this topic Ask about this topic

Covariate shift occurs when source and deployment inputs have different distributions, even though the rule linking inputs to labels is assumed to stay the same. Common approaches reweight labeled source examples using estimated density ratios, match source and target feature distributions, or learn representations that transfer better; clipping and regularization can limit unstable weights. Researchers also use conformal prediction, calibration, and risk-estimation methods to assess coverage and reliability without many labeled target examples, including in image, medical, and geographic transfer settings. A central tension is that reweighting needs adequate source–target overlap and can become noisy when ratios are large, while real shifts may also change the label relationship and invalidate covariate-only assumptions.

Latest papers

51 papers8 papers in the last four weeks, level with the four weeks before. 0.0% of all new papers.

Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift
May 20, 2026Jinzong Dong, Zhaohui Jiang, Bo YangProbability Calibration Unsupervised Domain Adaptation
School of Automation, Central South University, Changsha, China.
Anytime PAC-Bayes for Constrained Density-Ratio Networks under Covariate Shift
May 17, 2026Paulo Akira F. Enabe, Rodrigo ProvasiPAC-Bayesian Generalization Bounds Domain Adaptation
Escola Politénica University of São Paulo Department of Structural and Geotechnical Engineering
TILT: Target-induced loss tilting under covariate shift
May 14, 2026Kakei Yamamoto, Martin J. WainwrightUnsupervised Domain Adaptation Distribution Shift
Lab for Information and Decision Systems · Statistics and Data Science Center · EECS† +2
Post-hoc Selective Classification for Reliable Synthetic Image Detection
May 9, 2026Kaixiang Zheng, Jacob H. SeidmanUncertainty Quantification Distribution Shift
University of Waterloo · Reality Defender
Learning Large-Scale Modular Addition with an Auxiliary Modulus
May 8, 2026Hanato Kikuchi, Ryosuke Masuya, Kazuhiko Kawamoto +1Boolean Function Learning Distribution Shift
Chiba University · National Institute of Informatics
Jacobian-Velocity Bounds for Deployment Risk Under Covariate Drift
May 6, 2026Jonathan R. LandersDistribution Shift Robustness Distribution Shift
Independent Researcher · New York, NY, USA
Weight Clipping for Robust Conformal Inference under Unbounded Covariate Shifts
May 3, 2026James Wang, Surbhi GoelDistribution Shift Robustness Weighted Conformal Prediction
Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA 19104, USA.
CURE-OOD: Benchmarking Out-of-Distribution Detection for Survival Prediction
May 1, 2026Wenjie Zhao, Jia Li, Mingrui Liu +2Cancer Survival Prediction Distribution Shift
University of Texas at Dallas · George Mason University · UT Southwestern Medical Center
Multicalibration for Unbiased Model-Based Prevalence Estimation
Apr 23, 2026Fridolin Linder, Thomas Leeper, Daniel Haimovich +3Probability Calibration Distribution Shift
Meta Platforms Inc. · The London School of Economics and Political Science
ShapShift: Explaining Model Prediction Shifts with Subgroup Conditional Shapley Values
Apr 13, 2026Tom Bewley, Salim I. Amoukou, Emanuele Albini +2Shapley Value Attribution Feature Attribution
J.P. Morgan AI Research
Adversarial Label Invariant Graph Data Augmentations for Out-of-Distribution Generalization
Apr 9, 2026Simon Zhang, Ryan P. DeMilt, Kun Jin +1OOD Generalization Covariate Shift
Department of Computer Science, Purdue University, West Lafayette, USA · Department of Computer Science and Engineering, The Ohio State University, Columbus Ohio, USA · Department of Industrial and Systems Engineering, The Ohio State University, Columbus Ohio, USA
Conformal PM2.5 Mapping Under Spatial Covariate Shift: Satellite-Reanalysis Fusion for Africa's Green Industrial Transition
Apr 5, 2026Yaw Osei Adjei, Davis Opoku, Ephraim Abotsi +4Air Quality Prediction Covariate Shift
Kwame Nkrumah University of Science and Technology, Kumasi, Ghana · Dept. of Computer Science, Kwame Nkrumah Univ. of Science and Technology, Kumasi, Ghana
Transfer Learning for Loan Recovery Prediction under Distribution Shifts with Heterogeneous Feature Spaces
Apr 3, 2026Christopher Gerling, Hanqiu Peng, Ying Chen +1Cross-Domain Transfer Learning Domain Adaptation
Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany · National University of Singapore, 10 Lower Kent Ridge Road, Singapore 119076, Singapore
Transfer Learning for Meta-analysis Under Covariate Shift
Apr 3, 2026Zilong Wang, Ali Abdeen, Turgay AyerCausal Inference Individualized Treatment Effect Estimation
Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA
Unbounded Density Ratio Estimation and Its Application to Covariate Shift Adaptation
Mar 31, 2026Ren-Rui Liu, Jun Fan, Lei Shi +1Density Ratio Estimation Statistical Learning Theory
School of Mathematical Sciences, Zhejiang University, Hangzhou 310058, China · Department of Mathematics, Hong Kong Baptist University, Kowloon, Hong Kong · School of Mathematical Sciences and Shanghai, Key Laboratory for Contemporary Applied Mathematics, Fudan University, Shanghai 200433, China
Conformalized Signal Temporal Logic Inference under Covariate Shift
Mar 28, 2026Yixuan Wang, Danyang Li, Matthew Cleaveland +2Signal Temporal Logic Conformal Prediction
Mingyu Cai and Yixuan Wang are with Mechanical Engineering, University of California, Riverside, CA, 92521, USA. · Danyang Li and Roberto Tron are with Mechanical Engineering, Boston University. · Matthew Cleaveland is with MIT Lincoln Laboratory, Lexington, MA, 02421, USA
KMM-CP: Practical Conformal Prediction under Covariate Shift via Selective Kernel Mean Matching
Mar 27, 2026Siddhartha Laghuvarapu, Rohan Deb, Jimeng SunWeighted Conformal Prediction Covariate Shift
Siebel School of Computing and Data Science, University of Illinois Urbana-Champaign, IL, USA.
Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs
Mar 19, 2026Nathan Weill, Kaizheng WangUnsupervised Domain Adaptation Covariate Shift
Department of IEOR, Columbia University. · Department of IEOR and Data Science Institute, Columbia University.
Tracking the Discriminative Axis: Dual Prototypes for Test-Time OOD Detection Under Covariate Shift
Mar 16, 2026Wooseok Lee, Jin Mo Yang, Saewoong Bahk +1OOD Detection Covariate Shift
Seoul National University, Seoul, Korea
CaDrift: A Time-dependent Causal Generator of Drifting Data Streams
Feb 23, 2026Eduardo V. L. Barboza, Jean Paul Barddal, Robert Sabourin +1Synthetic Data Generation Synthetic Benchmark Generation
LIVIA, École de Technologie Supérieure (ÉTS), Université du Québec, Montreal, QC, Canada · Graduate Program in Informatics (PPGIa), Pontifícia Universidade Católica do Paraná (PUCPR), Curitiba, PR, Brazil
Coverage Guarantees for Pseudo-Calibrated Conformal Prediction under Distribution Shift
Feb 16, 2026Farbod Siahkali, Ashwin Verma, Vijay GuptaConformal Prediction Distribution Shift Robustness
Elmore Family School of Electrical and Computer Engineering, at Purdue University, West Lafayette, IN 47907 USA

May 20, 2026J/K move · Enter open · S save

Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift

Jinzong Dong, Zhaohui Jiang, Bo Yang

School of Automation, Central South University, Changsha, China.

Probability Calibration Unsupervised Domain Adaptation Covariate Shift Model Calibration Uncertainty Calibration

Open paper AskPDF arXiv

Confidence calibration for classification models is vital in safety-critical decision-making scenarios and has received extensive attention. General confidence calibration methods assume training and test data are independent and identically distributed, limiting their effectiveness under covariate shifts. Previous calibration methods under covariate shift struggle with class-wise or canonical calibrations and often rely on unstable importance weighting when density ratios are large or unbounded. Given the above limitations, this paper rethinks confidence calibration under covariate shifts. First, we derive a necessary and sufficient condition for confidence calibration under covariate shifts, named Expectation consistency condition, which reveals covariate shifts do not necessarily lead to uncalibrated confidence and provides a weaker condition for confidence calibration than global covariate distribution alignment. Then, utilizing Expectation consistency condition, this paper proposes an unsupervised domain adaptation loss to calibrate confidence of the target domain, named Expectation consistency loss (ECL), which is compatible with canonical calibration, class-wise calibration, and top-label calibration. Third, we prove that computing ECL loss has the same sample complexity as Expected Calibration Error (ECE) and provide a theoretically grounded mini-batch trainable scheme for ECL loss. Finally, we validate the effectiveness of our method on both simulated and real-world covariate shift datasets.