TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/Feature Generation for Long-tail Classification

Feature Generation for Long-tail Classification

Rahul Vigneswaran, Marc T. Law, Vineeth N. Balasubramanian, Makarand Tapaswi

2021-11-10Few-Shot LearningLong-tail LearningGeneral ClassificationClassificationimbalanced classification
PaperPDFCode(official)

Abstract

The visual world naturally exhibits an imbalance in the number of object or scene instances resulting in a \emph{long-tailed distribution}. This imbalance poses significant challenges for classification models based on deep learning. Oversampling instances of the tail classes attempts to solve this imbalance. However, the limited visual diversity results in a network with poor representation ability. A simple counter to this is decoupling the representation and classifier networks and using oversampling only to train the classifier. In this paper, instead of repeatedly re-sampling the same image (and thereby features), we explore a direction that attempts to generate meaningful features by estimating the tail category's distribution. Inspired by ideas from recent work on few-shot learning, we create calibrated distributions to sample additional features that are subsequently used to train the classifier. Through several experiments on the CIFAR-100-LT (long-tail) dataset with varying imbalance factors and on mini-ImageNet-LT (long-tail), we show the efficacy of our approach and establish a new state-of-the-art. We also present a qualitative analysis of generated features using t-SNE visualizations and analyze the nearest neighbors used to calibrate the tail class distributions. Our code is available at https://github.com/rahulvigneswaran/TailCalibX.

Results

TaskDatasetMetricValueModel
Image ClassificationCIFAR-100-LT (ρ=50)Error Rate49.1CBD+TailCalibX
Image ClassificationCIFAR-100-LT (ρ=10)Error Rate38.87CBD+TailCalibX
Image ClassificationCIFAR-100-LT (ρ=100)Error Rate53.41CBD+TailCalibX
Image Classificationmini-ImageNet-LTError Rate55.27TailCalibX
Few-Shot Image ClassificationCIFAR-100-LT (ρ=50)Error Rate49.1CBD+TailCalibX
Few-Shot Image ClassificationCIFAR-100-LT (ρ=10)Error Rate38.87CBD+TailCalibX
Few-Shot Image ClassificationCIFAR-100-LT (ρ=100)Error Rate53.41CBD+TailCalibX
Few-Shot Image Classificationmini-ImageNet-LTError Rate55.27TailCalibX
Generalized Few-Shot ClassificationCIFAR-100-LT (ρ=50)Error Rate49.1CBD+TailCalibX
Generalized Few-Shot ClassificationCIFAR-100-LT (ρ=10)Error Rate38.87CBD+TailCalibX
Generalized Few-Shot ClassificationCIFAR-100-LT (ρ=100)Error Rate53.41CBD+TailCalibX
Generalized Few-Shot Classificationmini-ImageNet-LTError Rate55.27TailCalibX
Long-tail LearningCIFAR-100-LT (ρ=50)Error Rate49.1CBD+TailCalibX
Long-tail LearningCIFAR-100-LT (ρ=10)Error Rate38.87CBD+TailCalibX
Long-tail LearningCIFAR-100-LT (ρ=100)Error Rate53.41CBD+TailCalibX
Long-tail Learningmini-ImageNet-LTError Rate55.27TailCalibX
Generalized Few-Shot LearningCIFAR-100-LT (ρ=50)Error Rate49.1CBD+TailCalibX
Generalized Few-Shot LearningCIFAR-100-LT (ρ=10)Error Rate38.87CBD+TailCalibX
Generalized Few-Shot LearningCIFAR-100-LT (ρ=100)Error Rate53.41CBD+TailCalibX
Generalized Few-Shot Learningmini-ImageNet-LTError Rate55.27TailCalibX

Related Papers

GLAD: Generalizable Tuning for Vision-Language Models2025-07-17Adversarial attacks to image classification systems using evolutionary algorithms2025-07-17Efficient Calisthenics Skills Classification through Foreground Instance Selection and Depth Estimation2025-07-16Safeguarding Federated Learning-based Road Condition Classification2025-07-16AI-Enhanced Pediatric Pneumonia Detection: A CNN-Based Approach Using Data Augmentation and Generative Adversarial Networks (GANs)2025-07-13Doodle Your Keypoints: Sketch-Based Few-Shot Keypoint Detection2025-07-10An Enhanced Privacy-preserving Federated Few-shot Learning Framework for Respiratory Disease Diagnosis2025-07-10Few-Shot Learning by Explicit Physics Integration: An Application to Groundwater Heat Transport2025-07-08