paper-with-me

홈 › Papers

Nuisance-Label Supervision: Robustness Improvement by Free Labels

2021-10-14 · Xinyue Wei, Weichao Qiu, Yi Zhang, Zihao Xiao, Alan Yuille

In this paper, we present a Nuisance-label Supervision (NLS) module, which can make models more robust to nuisance factor variations. Nuisance factors are those irrelevant to a task, and an ideal model should be invariant to them. For example, an activity recognition model should perform consistently regardless of the change of clothes and background. But our experiments show existing models are far from this capability. So we explicitly supervise a model with nuisance labels to make extracted features less dependent on nuisance factors. Although the values of nuisance factors are rarely annotated, we demonstrate that besides existing annotations, nuisance labels can be acquired freely from data augmentation and synthetic data. Experiments show consistent improvement in robustness towards image corruption and appearance change in action recognition.

📄 PDF Abstract BibTeX arXiv:2110.07118

Code (0)

등록된 구현이 없습니다.

Tasks

Action RecognitionActivity RecognitionData Augmentation

Similar Papers 제목 키워드 기반

RED-Sphere: Hyperspherical Residual Edge Debiasing for Cross-Population Fundus Disease Domain Generalization

2026-07-12 · Yan Lin, Ziheng Wang, Shuang Chen, Amir Atapour-Abarghouei 외 arxiv

Medical image classifiers are often trained within one source population, yet clinical deployment requires robustness to patients whose appearance, acquisition style, and disease prevalence differ from the source cohort.…

Domain GeneralizationImage Classification

A General Framework for Treatment Effect Estimation in Semi-Supervised and High Dimensional Settings

2022-01-03 · Abhishek Chakrabortty, Guorong Dai

In this article, we aim to provide a general and complete understanding of semi-supervised (SS) causal inference for treatment effects. Specifically, we consider two such estimands: (a) the average treatment effect and (…

Causal Inference

Robustness to Spurious Correlations Improves Semantic Out-of-Distribution Detection

2023-02-08 · Lily H. Zhang, Rajesh Ranganath

Methods which utilize the outputs or feature representations of predictive models have emerged as promising approaches for out-of-distribution (OOD) detection of image inputs. However, these methods struggle to detect OO…

Domain GeneralizationOut-of-Distribution DetectionOut of Distribution (OOD) Detection

Classification under Nuisance Parameters and Generalized Label Shift in Likelihood-Free Inference

2024-02-08 · Luca Masserano, Alex Shen, Michele Doro, Tommaso Dorigo 외

An open scientific challenge is how to classify events with reliable measures of uncertainty, when we have a mechanistic model of the data-generating process but the distribution over both labels and latent nuisance para…

Domain AdaptationUncertainty Quantificationvalid

Sensor-Conditioned Representation Learning via Scene-Relevant Observation Quotients

2026-06-15 · Yan Jiao, Pin-Han Ho, Limei Peng arxiv

Learned representations in intelligent sensing systems are often evaluated by reconstruction fidelity or downstream prediction accuracy, but these criteria do not specify which latent distinctions are justified by the se…

Representation Learning