用激活匹配生成简洁可信的模型决策解释。
Activation Matching for Explanation Generation
- 通过自编码器学习二值掩码，使解释图像保留原图分类结果和中间激活。
- 生成的掩码小而清晰，能忠实还原模型决策过程。
- 适合需要理解模型推理逻辑的研究者与开发者。
本文提出一种基于激活匹配的解释生成方法，针对预训练分类器在任意输入图像上的决策行为，生成最小且忠实的解释。给定输入图像 $x$ 和冻结模型 $f$，通过训练轻量级自编码器输出二值掩码 $m$，使得解释 $e = m igodot x$ 同时保留模型预测结果与中间层激活分布。目标函数包含：(i) 多层激活匹配，使用 KL 散度对齐激活分布，交叉熵保持顶层类别一致；(ii) 掩码先验——L1 面积约束以实现最小化，二值化惩罚项确保掩码为 0/1 值，总变差正则增强紧凑性；(iii) 反向推理约束以保障解释的忠实性与必要性。该方法生成的小尺寸、可读性强的掩码，能有效保留分类器行为并剔除无关区域，为模型决策提供实用且可信的极简解释。
原文摘要 · Abstract (English)
In this paper we introduce an activation-matching--based approach to generate minimal, faithful explanations for the decision-making of a pretrained classifier on any given image. Given an input image $x$ and a frozen model $f$, we train a lightweight autoencoder to output a binary mask $m$ such that the explanation $e = m \odot x$ preserves both the model's prediction and the intermediate activations of \(x\). Our objective combines: (i) multi-layer activation matching with KL divergence to align distributions and cross-entropy to retain the top-1 label for both the image and the explanation; (ii) mask priors -- L1 area for minimality, a binarization penalty for crisp 0/1 masks, and total variation for compactness; and (iii) abductive constraints for faithfulness and necessity. Together, these objectives yield small, human-interpretable masks that retain classifier behavior while discarding irrelevant input regions, providing practical and faithful minimalist explanations for the decision making of the underlying model.
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。