paper-with-me

홈 › Papers

Understanding Intrinsic Robustness Using Label Uncertainty

2021-07-07 · ICLR 2022 4 · Xiao Zhang, David Evans

A fundamental question in adversarial machine learning is whether a robust classifier exists for a given task. A line of research has made some progress towards this goal by studying the concentration of measure, but we argue standard concentration fails to fully characterize the intrinsic robustness of a classification problem since it ignores data labels which are essential to any classification task. Building on a novel definition of label uncertainty, we empirically demonstrate that error regions induced by state-of-the-art models tend to have much higher label uncertainty than randomly-selected subsets. This observation motivates us to adapt a concentration estimation algorithm to account for label uncertainty, resulting in more accurate intrinsic robustness measures for benchmark image classification problems.

📄 PDF Abstract BibTeX arXiv:2107.03250

Code (1)

xiaozhanguva/intrinsic_rob_lu 공식 구현 pytorch

Tasks

Adversarial RobustnessClassificationimage-classificationImage Classification

Similar Papers 제목 키워드 기반

Correlated Uncertainty for Learning Dense Correspondences from Noisy Labels

2019-12-01 · NeurIPS 2019 12 · Natalia Neverova, David Novotny, Andrea Vedaldi

Many machine learning methods depend on human supervision to achieve optimal performance. However, in tasks such as DensePose, where the goal is to establish dense visual correspondences between images, the quality of ma…

Visual Analysis of Prediction Uncertainty in Neural Networks for Deep Image Synthesis

2024-05-22 · Soumya Dutta, Faheem Nizar, Ahmad Amaan, Ayan Acharya

Ubiquitous applications of Deep neural networks (DNNs) in different artificial intelligence systems have led to their adoption in solving challenging visualization problems in recent years. While sophisticated DNNs offer…

DiversityImage GenerationPrediction

Learn to Estimate Labels Uncertainty for Quality Assurance

2019-09-17 · Agnieszka Tomczack, Nassir Navab, Shadi Albarqouni

Deep Learning sets the state-of-the-art in many challenging tasks showing outstanding performance in a broad range of applications. Despite its success, it still lacks robustness hindering its adoption in medical applica…

Bayesian InferenceDeep Learning

Delving Aleatoric Uncertainty in Medical Image Segmentation via Vision Foundation Models

2026-04-13 · Ruiyang Li, Fang Liu, Licheng Jiao, Xinglin Xie 외 arxiv

Medical image segmentation supports clinical workflows by precisely delineating anatomical structures and lesions. However, medical image datasets medical image datasets suffer from acquisition noise and annotation ambig…

Medical Image SegmentationTumor Segmentation

On the Robustness of Monte Carlo Dropout Trained with Noisy Labels

2021-03-22 · Purvi Goel, Li Chen

The memorization effect of deep learning hinders its performance to effectively generalize on test set when learning with noisy labels. Prior study has discovered that epistemic uncertainty techniques are robust when tra…

Learning with noisy labelsMemorization