SLT: Robust Quantum Neural Networks for Noisy-Label Medical Image Classification via Supermartingale-based Label Transition

2026-07-13

AI Summary

This work addresses the performance degradation of quantum neural networks in small-scale medical image classification caused by label noise. To tackle this issue, the authors propose SLT, an anchor-free loss correction framework that, for the first time, incorporates supermartingale theory into quantum learning with noisy labels. By modeling the entropy reduction of predictive distributions as a supermartingale, SLT leverages its monotonicity to dynamically refine the label transition matrix without relying on anchor points, thereby enabling robust training with guaranteed convergence stability. Experimental results demonstrate that SLT consistently outperforms existing classical methods across multiple public small-scale medical image datasets under both synthetic and real-world label noise scenarios.

Abstract

Noisy-label learning in small-scale medical image classification is challenging and hinders the superiority of deep neural networks. Recent studies suggest that quantum neural networks (QNNs) have shown potential in limited-data regimes, yet their use for noisy-label learning remains under-explored. A key obstacle is QNNs' intrinsic "natural smoothness", which may regularize training but also obscure high-confidence samples needed for noise-transition estimation. We propose Supermartingale-based Label Transition (SLT), an anchor-free loss correction framework for robust QNN-based medical image classification under noisy labels. SLT models entropy reduction in predictive distributions as a supermartingale and uses its monotonic behavior to identify stable transition-matrix refinement steps. This enables dynamic transition updates while reducing noise-driven oscillations during QNN training. We further provide a convergence analysis showing that the proposed transition-refinement process reaches a steady state. Experiments on multiple public small-scale medical image datasets demonstrate that SLT consistently improves QNN-based classification and stably outperforms classic noise-label learning baselines under synthetic and real-world label noise.

Problem

• noisy-label learning
• quantum neural networks
• medical image classification
• label noise
• small-scale datasets

Innovation

• Quantum Neural Networks
• Noisy-Label Learning
• Supermartingale
• Label Transition Matrix
• Medical Image Classification