潜在扩散反事实解释
Latent Diffusion Counterfactual Explanations
AI总结：
针对反事实生成中噪声梯度问题，提出基于潜在扩散模型的LDCE方法，利用共识引导机制过滤对抗性梯度，提升生成质量并洞察模型错误。
摘要：
反事实解释已成为阐明不透明黑盒模型行为的一种有前景的方法。近期，多项工作利用像素空间扩散模型进行反事实生成。为了处理反事实生成过程中出现的噪声、对抗性梯度——这些梯度会导致不真实的伪影或仅仅是对抗性扰动——这些方法要么需要辅助的对抗鲁棒模型，要么需要计算密集的引导方案。然而，这些要求限制了它们的适用性，例如在模型训练数据访问受限的场景中。为解决这些限制，我们引入了潜在扩散反事实解释（LDCE）。LDCE利用近期基于类别或文本条件的基础潜在扩散模型的能力，以加速反事实生成并聚焦于数据的重要语义部分。此外，我们提出了一种新颖的共识引导机制，以过滤掉与扩散模型隐式分类器不对齐的噪声、对抗性梯度。我们展示了LDCE在多种不同学习范式下训练于不同数据集上的模型中的广泛适用性。最后，我们展示了LDCE如何提供对模型错误的洞察，增强我们对黑盒模型行为的理解。
英文摘要：
Counterfactual explanations have emerged as a promising method for elucidating the behavior of opaque black-box models. Recently, several works leveraged pixel-space diffusion models for counterfactual generation. To handle noisy, adversarial gradients during counterfactual generation -- causing unrealistic artifacts or mere adversarial perturbations -- they required either auxiliary adversarially robust models or computationally intensive guidance schemes. However, such requirements limit their applicability, e.g., in scenarios with restricted access to the model's training data. To address these limitations, we introduce Latent Diffusion Counterfactual Explanations (LDCE). LDCE harnesses the capabilities of recent class- or text-conditional foundation latent diffusion models to expedite counterfactual generation and focus on the important, semantic parts of the data. Furthermore, we propose a novel consensus guidance mechanism to filter out noisy, adversarial gradients that are misaligned with the diffusion model's implicit classifier. We demonstrate the versatility of LDCE across a wide spectrum of models trained on diverse datasets with different learning paradigms. Finally, we showcase how LDCE can provide insights into model errors, enhancing our understanding of black-box model behavior.
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。
