No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problems

• URL: http://arxiv.org/abs/2011.12945v2
• Date: Sun, 10 Apr 2022 23:01:14 GMT
• Title: No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problems
• Authors: Nimit S. Sohoni, Jared A. Dunnmon, Geoffrey Angus, Albert Gu, Christopher R\'e
• Abstract summary: In real-world classification tasks, each class often comprises multiple finer-grained "subclasses" As the subclass labels are frequently unavailable, models trained using only the coarser-grained class labels often exhibit highly variable performance across different subclasses. We propose GEORGE, a method to both measure and mitigate hidden stratification even when subclass labels are unknown.
• Score: 20.253644336965042
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In real-world classification tasks, each class often comprises multiple finer-grained "subclasses." As the subclass labels are frequently unavailable, models trained using only the coarser-grained class labels often exhibit highly variable performance across different subclasses. This phenomenon, known as hidden stratification, has important consequences for models deployed in safety-critical applications such as medicine. We propose GEORGE, a method to both measure and mitigate hidden stratification even when subclass labels are unknown. We first observe that unlabeled subclasses are often separable in the feature space of deep neural networks, and exploit this fact to estimate subclass labels for the training data via clustering techniques. We then use these approximate subclass labels as a form of noisy supervision in a distributionally robust optimization objective. We theoretically characterize the performance of GEORGE in terms of the worst-case generalization error across any subclass. We empirically validate GEORGE on a mix of real-world and benchmark image classification datasets, and show that our approach boosts worst-case subclass accuracy by up to 22 percentage points compared to standard training techniques, without requiring any prior information about the subclasses.

Related papers

• Exploration of Class Center for Fine-Grained Visual Classification [7.120809788357261]
  Fine-grained visual classification is a challenging task due to intra-class variances and subtle inter-class differences. We propose a loss function named exploration of class center, which consists of a multiple class-center constraint and a class-center label generation. Our method can be easily integrated with existing fine-grained visual classification approaches as a loss function, to further boost excellent performance with only slight training costs.
  arXiv  Detail & Related papers  (2024-07-05T04:11:09Z)
• Embracing Diversity: Interpretable Zero-shot classification beyond one vector per class [16.101460010750458]
  We argue that to represent diversity within a class, zero-shot classification should move beyond a single vector. We propose a method to encode and account for diversity within a class using inferred attributes, still in the zero-shot setting without retraining. We find our method consistently outperforms standard zero-shot classification over a large suite of datasets.
  arXiv  Detail & Related papers  (2024-04-25T16:29:06Z)
• Active Generalized Category Discovery [60.69060965936214]
  Generalized Category Discovery (GCD) endeavors to cluster unlabeled samples from both novel and old classes. We take the spirit of active learning and propose a new setting called Active Generalized Category Discovery (AGCD) Our method achieves state-of-the-art performance on both generic and fine-grained datasets.
  arXiv  Detail & Related papers  (2024-03-07T07:12:24Z)
• Exploring Vacant Classes in Label-Skewed Federated Learning [113.65301899666645]
  Label skews, characterized by disparities in local label distribution across clients, pose a significant challenge in federated learning. This paper introduces FedVLS, a novel approach to label-skewed federated learning that integrates vacant-class distillation and logit suppression simultaneously.
  arXiv  Detail & Related papers  (2024-01-04T16:06:31Z)
• Subclass-balancing Contrastive Learning for Long-tailed Recognition [38.31221755013738]
  Long-tailed recognition with imbalanced class distribution naturally emerges in practical machine learning applications. We propose a novel subclass-balancing contrastive learning'' approach that clusters each head class into multiple subclasses of similar sizes as the tail classes. We evaluate SBCL over a list of long-tailed benchmark datasets and it achieves the state-of-the-art performance.
  arXiv  Detail & Related papers  (2023-06-28T05:08:43Z)
• Parametric Classification for Generalized Category Discovery: A Baseline Study [70.73212959385387]
  Generalized Category Discovery (GCD) aims to discover novel categories in unlabelled datasets using knowledge learned from labelled samples. We investigate the failure of parametric classifiers, verify the effectiveness of previous design choices when high-quality supervision is available, and identify unreliable pseudo-labels as a key problem. We propose a simple yet effective parametric classification method that benefits from entropy regularisation, achieves state-of-the-art performance on multiple GCD benchmarks and shows strong robustness to unknown class numbers.
  arXiv  Detail & Related papers  (2022-11-21T18:47:11Z)
• Latent Preserving Generative Adversarial Network for Imbalance classification [17.992830267031877]
  We present an end-to-end deep generative classifier. In this paper, we propose a domain-constraint autoencoder to preserve the latent-space as prior for a generator. Experiments are carried out on three different multi-class imbalanced problems and a comparison with state-of-the-art methods.
  arXiv  Detail & Related papers  (2022-09-04T07:49:27Z)
• On Non-Random Missing Labels in Semi-Supervised Learning [114.62655062520425]
  Semi-Supervised Learning (SSL) is fundamentally a missing label problem. We explicitly incorporate "class" into SSL. Our method not only significantly outperforms existing baselines but also surpasses other label bias removal SSL methods.
  arXiv  Detail & Related papers  (2022-06-29T22:01:29Z)
• Prototypical Classifier for Robust Class-Imbalanced Learning [64.96088324684683]
  We propose textitPrototypical, which does not require fitting additional parameters given the embedding network. Prototypical produces balanced and comparable predictions for all classes even though the training set is class-imbalanced. We test our method on CIFAR-10LT, CIFAR-100LT and Webvision datasets, observing that Prototypical obtains substaintial improvements compared with state of the arts.
  arXiv  Detail & Related papers  (2021-10-22T01:55:01Z)
• Learning and Evaluating Representations for Deep One-class Classification [59.095144932794646]
  We present a two-stage framework for deep one-class classification. We first learn self-supervised representations from one-class data, and then build one-class classifiers on learned representations. In experiments, we demonstrate state-of-the-art performance on visual domain one-class classification benchmarks.
  arXiv  Detail & Related papers  (2020-11-04T23:33:41Z)
• Beyond cross-entropy: learning highly separable feature distributions for robust and accurate classification [22.806324361016863]
  We propose a novel approach for training deep robust multiclass classifiers that provides adversarial robustness. We show that the regularization of the latent space based on our approach yields excellent classification accuracy.
  arXiv  Detail & Related papers  (2020-10-29T11:15:17Z)

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.