Robust Label Shift Quantification

• URL: http://arxiv.org/abs/2502.03174v1
• Date: Wed, 05 Feb 2025 13:51:24 GMT
• Title: Robust Label Shift Quantification
• Authors: Alexandre Lecestre,
• Abstract summary: We propose robust estimators of the label distribution which turn out to coincide with the Maximum Likelihood Estimator. Our results provide theoretical validation for empirical observations on the robustness of Maximum Likelihood Label Shift.
• Score: 55.2480439325792
• License:
• Abstract: In this paper, we investigate the label shift quantification problem. We propose robust estimators of the label distribution which turn out to coincide with the Maximum Likelihood Estimator. We analyze the theoretical aspects and derive deviation bounds for the proposed method, providing optimal guarantees in the well-specified case, along with notable robustness properties against outliers and contamination. Our results provide theoretical validation for empirical observations on the robustness of Maximum Likelihood Label Shift.

Related papers

• Federated Generalised Variational Inference: A Robust Probabilistic Federated Learning Framework [12.454538785810259]
  FedGVI is a probabilistic Federated Learning (FL) framework that is provably robust to both prior and likelihood misspecification. We offer theoretical analysis in terms of fixed-point convergence, optimality of the cavity distribution, and provable robustness.
  arXiv  Detail & Related papers  (2025-02-02T16:39:37Z)
• Learning from Noisy Labels via Conditional Distributionally Robust Optimization [5.85767711644773]
  crowdsourcing has emerged as a practical solution for labeling large datasets. It presents a significant challenge in learning accurate models due to noisy labels from annotators with varying levels of expertise.
  arXiv  Detail & Related papers  (2024-11-26T05:03:26Z)
• Label Shift Quantification with Robustness Guarantees via Distribution Feature Matching [3.2013172123155615]
  We first present a unifying framework, distribution feature matching (DFM), that recovers as particular instances various estimators introduced in previous literature. We then extend this analysis to study robustness of DFM procedures in the misspecified setting under departure from the exact label shift hypothesis. These theoretical findings are confirmed by a detailed numerical study on simulated and real-world datasets.
  arXiv  Detail & Related papers  (2023-06-07T12:17:34Z)
• Variational Classification [51.2541371924591]
  We derive a variational objective to train the model, analogous to the evidence lower bound (ELBO) used to train variational auto-encoders. Treating inputs to the softmax layer as samples of a latent variable, our abstracted perspective reveals a potential inconsistency. We induce a chosen latent distribution, instead of the implicit assumption found in a standard softmax layer.
  arXiv  Detail & Related papers  (2023-05-17T17:47:19Z)
• Regression with Label Differential Privacy [64.21020761920322]
  We derive a label DP randomization mechanism that is optimal under a given regression loss function. We prove that the optimal mechanism takes the form of a "randomized response on bins"
  arXiv  Detail & Related papers  (2022-12-12T17:41:32Z)
• Maximum Likelihood Uncertainty Estimation: Robustness to Outliers [3.673994921516517]
  Outliers or noisy labels in training data results in degraded performances as well as incorrect estimation of uncertainty. We propose the use of a heavy-tailed distribution to improve the robustness to outliers.
  arXiv  Detail & Related papers  (2022-02-03T10:41:34Z)
• Optimal variance-reduced stochastic approximation in Banach spaces [114.8734960258221]
  We study the problem of estimating the fixed point of a contractive operator defined on a separable Banach space. We establish non-asymptotic bounds for both the operator defect and the estimation error.
  arXiv  Detail & Related papers  (2022-01-21T02:46:57Z)
• A Unified Joint Maximum Mean Discrepancy for Domain Adaptation [73.44809425486767]
  This paper theoretically derives a unified form of JMMD that is easy to optimize. From the revealed unified JMMD, we illustrate that JMMD degrades the feature-label dependence that benefits to classification. We propose a novel MMD matrix to promote the dependence, and devise a novel label kernel that is robust to label distribution shift.
  arXiv  Detail & Related papers  (2021-01-25T09:46:14Z)
• Exploiting Sample Uncertainty for Domain Adaptive Person Re-Identification [137.9939571408506]
  We estimate and exploit the credibility of the assigned pseudo-label of each sample to alleviate the influence of noisy labels. Our uncertainty-guided optimization brings significant improvement and achieves the state-of-the-art performance on benchmark datasets.
  arXiv  Detail & Related papers  (2020-12-16T04:09:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.