用概念瓶颈模型提升黑盒模型可解释性并改进性能。
Intervening in Black Box: Concept Bottleneck Model for Enhancing Human Neural Network Mutual Understanding
- 通过全局梯度识别有害概念并优化模型结构。
- 在多个数据集上最高提升2.64%准确率，平均增益1.03%。
- 适合关注模型可解释性与性能平衡的研究者。
深度学习模型层数更深、参数更多，导致决策过程难以理解。现有解释方法大多缺乏有效干预或仅限样本层面，无法修改模型本身。为此，我们提出概念瓶颈模型增强人机互信（CBM-HNMU），利用概念瓶颈模型（CBM）作为可解释框架，近似黑盒推理并传达概念理解。基于全局梯度贡献，自动识别并精炼有害概念（移除或替换）。随后将修正后的知识蒸馏回黑盒模型，同时提升可解释性与准确率。我们在Flower-102、CIFAR-10、CIFAR-100、FGVC-Aircraft和CUB-200等数据集上评估了基于CNN和Transformer的多种模型，最高实现2.64%的准确率提升，平均准确率提升1.03%。源代码已公开：https://github.com/XiGuaBo/CBM-HNMU。
原文摘要 · Abstract (English)
Recent advances in deep learning have led to increasingly complex models with deeper layers and more parameters, reducing interpretability and making their decisions harder to understand. While many methods explain black-box reasoning, most lack effective interventions or only operate at sample-level without modifying the model itself. To address this, we propose the Concept Bottleneck Model for Enhancing Human-Neural Network Mutual Understanding (CBM-HNMU). CBM-HNMU leverages the Concept Bottleneck Model (CBM) as an interpretable framework to approximate black-box reasoning and communicate conceptual understanding. Detrimental concepts are automatically identified and refined (removed/replaced) based on global gradient contributions. The modified CBM then distills corrected knowledge back into the black-box model, enhancing both interpretability and accuracy. We evaluate CBM-HNMU on various CNN and transformer-based models across Flower-102, CIFAR-10, CIFAR-100, FGVC-Aircraft, and CUB-200, achieving a maximum accuracy improvement of 2.64% and a maximum increase in average accuracy across 1.03%. Source code is available at: https://github.com/XiGuaBo/CBM-HNMU.
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。