Class-Specific Branch Attention for Mitigating Gradient Interference under Class Imbalance

📅 2026-06-04
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the suppression of minority-class learning in deep neural networks under severe class imbalance, which arises from shared feature representations. To mitigate this issue, the authors propose a Class-Specific Branch Attention (CSBA) mechanism. By analyzing inter-layer gradient flows and constructing a gradient conflict matrix based on class-specific gradient cosine similarity, they reveal—through the lens of optimization dynamics—for the first time how majority classes dominate and suppress gradients of minority classes. A lightweight channel reweighting module is integrated into a multi-branch convolutional architecture to implicitly decouple features and gradients in a class-aware manner. Experiments demonstrate that the proposed method significantly improves minority-class performance without compromising overall accuracy: the F1 score for the Physical-Damage class increases from 0.261 to 0.522, and Macro-F1 on CIFAR-10-LT rises from 0.595 to 0.655.
📝 Abstract
Deep neural networks trained under severe class imbalance often exhibit degraded performance, typically attributed to statistical bias. In this work, we identify a complementary optimization-level pathology: inter-class gradient interference within shared representations, where gradients from majority classes suppress minority-class learning. To analyze this phenomenon, we introduce a diagnostic framework based on layer-wise gradient flow analysis and a Gradient Conflict Matrix, which quantifies interference using cosine similarity between class-specific gradients. Using this framework, we study multi-branch convolutional architectures and propose a lightweight modification, Class-Specific Branch Attention (CSBA), that enables branch-specific channel reweighting to reduce gradient coupling. This mechanism promotes implicit feature decoupling across branches while preserving architectural simplicity. Empirically, CSBA improves minority-class performance, increasing the F1 score for the Physical-Damage class from 0.261 to 0.522 under severe imbalance, while maintaining comparable overall accuracy. Validation on CIFAR-10-LT confirms that this behavior generalizes across imbalanced visual recognition settings, with Macro-F1 improving from 0.595 to 0.655. More broadly, our findings highlight the importance of considering optimization dynamics alongside statistical methods when designing architectures for imbalanced learning.
Problem

Research questions and friction points this paper is trying to address.

class imbalance
gradient interference
shared representations
minority-class learning
optimization dynamics
Innovation

Methods, ideas, or system contributions that make the work stand out.

gradient interference
class imbalance
Class-Specific Branch Attention
feature decoupling
imbalanced learning
🔎 Similar Papers
No similar papers found.
A
Arush Singhal
Thapar Institute of Engineering and Technology, Patiala, India
U
Umang Soni
Netaji Subhash University of Technology, New Delhi, India