Machine Unlearning for Robust DNNs: Attribution-Guided Partitioning and Neuron Pruning in Noisy Environments

TL;DR

This paper introduces a novel machine unlearning framework that uses attribution-guided data partitioning and neuron pruning to improve the robustness of deep neural networks against noisy data, without extensive retraining.

Contribution

It proposes an integrated approach combining attribution-based data filtering, neuron pruning, and targeted fine-tuning to enhance noise robustness in DNNs, avoiding full retraining and explicit noise modeling.

Findings

Achieves about 10% accuracy improvement on CIFAR-10 with label noise.

Reduces retraining time by up to 47%.

Demonstrates scalability and effectiveness across tasks.

Abstract

Deep neural networks (DNNs) have achieved remarkable success across diverse domains, but their performance can be severely degraded by noisy or corrupted training data. Conventional noise mitigation methods often rely on explicit assumptions about noise distributions or require extensive retraining, which can be impractical for large-scale models. Inspired by the principles of machine unlearning, we propose a novel framework that integrates attribution-guided data partitioning, discriminative neuron pruning, and targeted fine-tuning to mitigate the impact of noisy samples. Our approach employs gradient-based attribution to probabilistically distinguish high-quality examples from potentially corrupted ones without imposing restrictive assumptions on the noise. It then applies regression-based sensitivity analysis to identify and prune neurons that are most vulnerable to noise. Finally, the resulting network is fine-tuned on the high-quality data subset to efficiently recover and enhance its generalization performance. This integrated unlearning-inspired framework provides several advantages over conventional noise-robust learning approaches. Notably, it combines data-level unlearning with model-level adaptation, thereby avoiding the need for full model retraining or explicit noise modeling. We evaluate our method on representative tasks (e.g., CIFAR-10 image classification and speech recognition) under various noise levels and observe substantial gains in both accuracy and efficiency. For example, our framework achieves approximately a 10% absolute accuracy improvement over standard retraining on CIFAR-10 with injected label noise, while reducing retraining time by up to 47% in some settings. These results demonstrate the effectiveness and scalability of the proposed approach for achieving robust generalization in noisy environments.