The prevalence of employing attention mechanisms has brought along concerns on the interpretability of attention distributions. Although it provides insights about how a model is operating, utilizing attention as the explanation of model predictions is still highly dubious. The community is still seeking more interpretable strategies for better identifying local active regions that contribute the most to the final decision. To improve the interpretability of existing attention models, we propose a novel Bilinear Representative Non-Parametric Attention (BR-NPA) strategy that captures the task-relevant human-interpretable information. The target model is first distilled to have higher-resolution intermediate feature maps. From which, representative features are then grouped based on local pairwise feature similarity, to produce finer-grained, more precise attention maps highlighting task-relevant parts of the input. The obtained attention maps are ranked according to the activity level of the compound feature, which provides information regarding the important level of the highlighted regions. The proposed model can be easily adapted in a wide variety of modern deep models, where classification is involved. Extensive quantitative and qualitative experiments showcase more comprehensive and accurate visual explanations compared to state-of-the-art attention models and visualizations methods across multiple tasks including fine-grained image classification, few-shot classification, and person re-identification, without compromising the classification accuracy. The proposed visualization model sheds imperative light on how neural networks `pay their attention' differently in different tasks.
翻译:利用关注机制的普遍程度使人们对关注分布的可解释性产生了关切,虽然它使人们对模型的运作方式有了深刻的认识,但利用模型预测模型的解释仍然非常可疑;社区仍在寻求更可解释的战略,以更好地确定对最终决定贡献最大的地方活跃区域;为了改进现有关注模式的可解释性,我们提议了一个新颖的双线代表非定位关注(BR-NPA)战略,以捕捉与任务相关的人类解释信息;目标模型首先蒸发,以获得更高分辨率的中间特征图。从中,然后根据当地对称特征的相似性将代表性特征分组,以制作精细的、更精确的注意地图,突出与任务相关的部分;为了改进现有关注模式的可理解性,我们提出了关于突出区域重要程度的信息;提议的模型很容易在涉及分类的多种现代深度模型中进行调整;广泛的定量和定性实验,展示了与州级的对等特征特征特征特征相似的更全面和准确的直观解释,从而在不以不同视角的图像分类中,在不同的图像分类中,包括拟议的对不同的图像分类中,对不同的视觉分类,对不同的图像进行细致的分类。