通过变量分解分析卷积网络对图像增强的敏感性与专业化
Exploring specialization and sensitivity of convolutional neural networks in the context of simultaneous image augmentations
- 用Sobol指数和Shapley值分解各增强操作对激活变化的影响
- 发现不同增强方式对网络内部响应的贡献差异显著
- 适合研究模型可解释性与生物神经网络的类比分析
受生物神经网络研究启发，本研究引入处理-对照范式，并通过多参数输入扰动丰富可解释人工智能方法。提出一种框架，用于探究输入数据增强对网络内部推理过程的影响。网络运行的内部变化通过激活方差体现，该方差可分解为各增强操作对应的分量，利用Sobol指数和Shapley值实现。这些指标可用于可视化对不同变量的敏感性，并指导激活掩码。此外，引入单类敏感性分析，根据目标破坏激活后产生的预测偏差筛选候选特征。基于观察到的类比，认为该框架或可迁移至复杂环境下生物神经网络的研究。
原文摘要 · Abstract (English)
Drawing parallels with the way biological networks are studied, we adapt the treatment--control paradigm to explainable artificial intelligence research and enrich it through multi-parametric input alterations. In this study, we propose a framework for investigating the internal inference impacted by input data augmentations. The internal changes in network operation are reflected in activation changes measured by variance, which can be decomposed into components related to each augmentation, employing Sobol indices and Shapley values. These quantities enable one to visualize sensitivity to different variables and use them for guided masking of activations. In addition, we introduce a way of single-class sensitivity analysis where the candidates are filtered according to their matching to prediction bias generated by targeted damaging of the activations. Relying on the observed parallels, we assume that the developed framework can potentially be transferred to studying biological neural networks in complex environments.
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。