DiG-IN：用于研究网络的扩散引导——揭示分类器差异、神经元可视化与视觉反事实解释
DiG-IN: Diffusion Guidance for Investigating Networks -- Uncovering Classifier Differences Neuron Visualisations and Visual Counterfactual Explanations
AI总结：
该研究提出DiG-IN扩散引导框架，通过生成优化分类器目标的图像，开展视觉反事实解释、分类器差异分析与神经元可视化，验证已有结论并发现新失效模式，且VCEs性能更优、通用性更强。
摘要：
尽管深度学习在ImageNet等复杂图像分类任务中取得了巨大进展，但由虚假特征等引发的意外失效模式，让人们对这些分类器在真实场景中的工作可靠性产生了质疑。此外，对于安全关键型任务，其决策的黑箱性质存在问题，迫切需要解释方法，或至少是能让决策具备合理性的方法。本文中，我们通过一个引导式图像生成框架，生成能优化由分类器导出的目标函数的图像，以解决这些问题。我们通过视觉反事实解释（VCEs）分析图像分类器的决策，通过分析分类器分歧最大的图像来检测系统性错误，还对神经元和虚假特征进行可视化。通过这种方式，我们验证了已有观察结果，例如对抗鲁棒模型的形状偏差，以及新的失效模式，例如零样本CLIP分类器的系统性误差。此外，我们的VCEs性能优于此前的工作，且通用性更强。
英文摘要：
While deep learning has led to huge progress in complex image classification tasks like ImageNet, unexpected failure modes, e.g. via spurious features, call into question how reliably these classifiers work in the wild. Furthermore, for safety-critical tasks the black-box nature of their decisions is problematic, and explanations or at least methods which make decisions plausible are needed urgently. In this paper, we address these problems by generating images that optimize a classifier-derived objective using a framework for guided image generation. We analyze the decisions of image classifiers by visual counterfactual explanations (VCEs), detection of systematic mistakes by analyzing images where classifiers maximally disagree, and visualization of neurons and spurious features. In this way, we validate existing observations, e.g. the shape bias of adversarially robust models, as well as novel failure modes, e.g. systematic errors of zero-shot CLIP classifiers. Moreover, our VCEs outperform previous work while being more versatile.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
