将神经启发的局部可塑性与度量学习结合，提升CNN的准确率和可解释性。
NM-Hebb: Coupling Local Hebbian Plasticity with Metric Learning for More Accurate and Interpretable CNNs
- 引入赫布正则化与可学习调制器，增强特征的结构化与可复用性。
- 在多个数据集上准确率提升2.0-10.0个百分点，类内距离压缩、类间间隔扩大。
- 适合对模型可解释性与资源效率有要求的应用场景，如医疗与自动驾驶。
深度卷积神经网络（CNN）虽精度高，但依赖全局梯度优化，易导致过拟合、冗余滤波器及可解释性差。为此，我们提出NM-Hebb，一种两阶段训练框架，融合神经启发的局部可塑性与距离感知监督。第一阶段在标准监督训练基础上，联合优化交叉熵损失与两个生物启发机制：(i) 赫布正则化，使激活的空间均值与对应卷积核权重均值对齐，促进结构化、可复用的特征基元；(ii) 可学习的神经调制器，调控弹性权重式巩固损失，保留有益参数而不冻结网络。第二阶段通过成对度量学习损失微调主干网络，显式压缩类内距离、扩大类间间隔。在CIFAR-10、CIFAR-100和TinyImageNet上，使用五种主干网络（ResNet-18、VGG-11、MobileNet-v2、EfficientNet-V2、DenseNet-121）测试，NM-Hebb相较基线和其他方法均有稳定提升：CIFAR-10 Top-1准确率提升+2.0–10.0个百分点，CIFAR-100提升+2.0–9.0个百分点，TinyImageNet提升最高达+4.3–8.9个百分点，归一化互信息（NMI）提升最高+0.15。定性可视化与滤波器级分析表明，该方法生成更结构化、更聚焦的特征，形成更紧凑、可解释的类别簇。整体而言，局部赫布可塑性与度量学习结合，使CNN兼具更高精度与更强可解释性，适用于资源受限与安全关键的AI部署。
原文摘要 · Abstract (English)
Deep Convolutional Neural Networks (CNNs) achieve high accuracy but often rely on purely global, gradient-based optimisation, which can lead to overfitting, redundant filters, and reduced interpretability. To address these limitations, we propose NM-Hebb, a two-phase training framework that integrates neuro-inspired local plasticity with distance-aware supervision. Phase 1 extends standard supervised training by jointly optimising a cross-entropy objective with two biologically inspired mechanisms: (i) a Hebbian regulariser that aligns the spatial mean of activations with the mean of the corresponding convolutional filter weights, encouraging structured, reusable primitives; and (ii) a learnable neuromodulator that gates an elastic-weight-style consolidation loss, preserving beneficial parameters without freezing the network. Phase 2 fine-tunes the backbone with a pairwise metric-learning loss, explicitly compressing intra-class distances and enlarging inter-class margins in the embedding space. Evaluated on CIFAR-10, CIFAR-100, and TinyImageNet across five backbones (ResNet-18, VGG-11, MobileNet-v2, EfficientNet-V2, DenseNet-121), NM-Hebb achieves consistent gains over baseline and other methods: Top-1 accuracy improves by +2.0-10.0 pp (CIFAR-10), +2.0-9.0 pp (CIFAR-100), and up to +4.3-8.9 pp (TinyImageNet), with Normalised Mutual Information (NMI) increased by up to +0.15. Qualitative visualisations and filter-level analyses further confirm that NM-Hebb produces more structured and selective features, yielding tighter and more interpretable class clusters. Overall, coupling local Hebbian plasticity with metric-based fine-tuning yields CNNs that are not only more accurate but also more interpretable, offering practical benefits for resource-constrained and safety-critical AI deployments.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。