Max-margin Class Imbalanced Learning with Gaussian Affinity

Abstract

Real-world object classes appear in imbalanced ratios. This poses a significant challenge for classifiers which get biased towards frequent classes. We hypothesize that improving the generalization capability of a classifier should improve learning on imbalanced datasets. Here, we introduce the first hybrid loss function that jointly performs classification and clustering in a single formulation. Our approach is based on an `affinity measure' in Euclidean space that leads to the following benefits: (1) direct enforcement of maximum margin constraints on classification boundaries, (2) a tractable way to ensure uniformly spaced and equidistant cluster centers, (3) flexibility to learn multiple class prototypes to support diversity and discriminability in feature space. Our extensive experiments demonstrate the significant performance improvements on visual classification and verification tasks on multiple imbalanced datasets. The proposed loss can easily be plugged in any deep architecture as a differentiable block and demonstrates robustness against different levels of data imbalance and corrupted labels.

References in corpus (9)

• SMOTE: Synthetic Minority Over-sampling Technique
• Prototypical Networks for Few-shot Learning
• Deep Learning Face Representation by Joint Identification-Verification
• Matching Networks for One Shot Learning
• Targeting Ultimate Accuracy: Face Recognition via Deep Embedding
• Batch-normalized Maxout Network in Network
• Do We Really Need to Collect Millions of Faces for Effective Face Recognition?
• Feature Transfer Learning for Deep Face Recognition with Under-Represented Data
• Large Margin Deep Networks for Classification

Cited by in corpus (8)

• Decoupling Representation and Classifier for Long-Tailed Recognition
• Striking the Right Balance with Uncertainty
• Rethinking Class-Balanced Methods for Long-Tailed Visual Recognition from a Domain Adaptation Perspective
• Jointly De-biasing Face Recognition and Demographic Attribute Estimation
• Exploring Optimal Deep Learning Models for Image-based Malware Variant Classification
• A Closer Look at Prototype Classifier for Few-shot Image Classification
• Incremental Cross-Domain Adaptation for Robust Retinopathy Screening via Bayesian Deep Learning
• On-target Adaptation