研究如何高效且有效地用解释引导模型
Studying How to Efficiently and Effectively Guide Models with Explanations
AI总结:
本研究系统评估了多种模型引导设计选择,提出基于边界框的高效引导方法及能量损失,能提升模型对物体特征的关注并改善分布偏移下的泛化。
摘要:
尽管深度神经网络性能卓越,但它们可能会基于与给定标签虚假相关的特征做出决策,从而损害泛化能力。为缓解这一问题,“模型引导”近年来受到广泛关注,即通过对模型的解释进行正则化,确保模型“因正确的理由而正确”。虽然已有多种实现模型引导的技术被提出,但迄今为止,这些方法的实验验证仅限于相对简单和/或合成的数据集。为了更好地理解在模型引导背景下探索的各种设计选择的有效性,本研究在PASCAL VOC 2007和MS COCO 2014数据集上,对各种损失函数、归因方法、模型和“引导深度”进行了深入评估。由于模型引导的标注成本可能限制其适用性,我们还特别关注效率。具体而言,我们通过边界框标注来引导模型,这比常用的分割掩码获取成本低得多,并评估了模型引导在有限(例如,仅1%的标注图像)或过于粗糙的标注下的鲁棒性。此外,我们提出使用EPG分数作为额外的评估指标和损失函数(“能量损失”)。我们表明,尽管仅使用包含背景区域的边界框标注,优化能量损失仍能使模型表现出对物体特定特征的明显关注。最后,我们表明这种模型引导可以改善分布偏移下的泛化能力。代码可在 https://github.com/sukrutrao/Model-Guidance 获取。
英文摘要:
Despite being highly performant, deep neural networks might base their decisions on features that spuriously correlate with the provided labels, thus hurting generalization. To mitigate this, 'model guidance' has recently gained popularity, i.e. the idea of regularizing the models' explanations to ensure that they are "right for the right reasons". While various techniques to achieve such model guidance have been proposed, experimental validation of these approaches has thus far been limited to relatively simple and / or synthetic datasets. To better understand the effectiveness of the various design choices that have been explored in the context of model guidance, in this work we conduct an in-depth evaluation across various loss functions, attribution methods, models, and 'guidance depths' on the PASCAL VOC 2007 and MS COCO 2014 datasets. As annotation costs for model guidance can limit its applicability, we also place a particular focus on efficiency. Specifically, we guide the models via bounding box annotations, which are much cheaper to obtain than the commonly used segmentation masks, and evaluate the robustness of model guidance under limited (e.g. with only 1% of annotated images) or overly coarse annotations. Further, we propose using the EPG score as an additional evaluation metric and loss function ('Energy loss'). We show that optimizing for the Energy loss leads to models that exhibit a distinct focus on object-specific features, despite only using bounding box annotations that also include background regions. Lastly, we show that such model guidance can improve generalization under distribution shifts. Code available at: https://github.com/sukrutrao/Model-Guidance.
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。