Deploying reliable deep learning techniques in interdisciplinary applications needs learned models to output accurate and ({even more importantly}) explainable predictions. Existing approaches typically explicate network outputs in a post-hoc fashion, under an implicit assumption that faithful explanations come from accurate predictions/classifications. We have an opposite claim that explanations boost (or even determine) classification. That is, end-to-end learning of explanation factors to augment discriminative representation extraction could be a more intuitive strategy to inversely assure fine-grained explainability, e.g., in those neuroimaging and neuroscience studies with high-dimensional data containing noisy, redundant, and task-irrelevant information. In this paper, we propose such an explainable geometric deep network dubbed as NeuroExplainer, with applications to uncover altered infant cortical development patterns associated with preterm birth. Given fundamental cortical attributes as network input, our NeuroExplainer adopts a hierarchical attention-decoding framework to learn fine-grained attentions and respective discriminative representations to accurately recognize preterm infants from term-born infants at term-equivalent age. NeuroExplainer learns the hierarchical attention-decoding modules under subject-level weak supervision coupled with targeted regularizers deduced from domain knowledge regarding brain development. These prior-guided constraints implicitly maximizes the explainability metrics (i.e., fidelity, sparsity, and stability) in network training, driving the learned network to output detailed explanations and accurate classifications. Experimental results on the public dHCP benchmark suggest that NeuroExplainer led to quantitatively reliable explanation results that are qualitatively consistent with representative neuroimaging studies.
translated by 谷歌翻译
Out-Of-Distribution (OOD) detection has received broad attention over the years, aiming to ensure the reliability and safety of deep neural networks (DNNs) in real-world scenarios by rejecting incorrect predictions. However, we notice a discrepancy between the conventional evaluation vs. the essential purpose of OOD detection. On the one hand, the conventional evaluation exclusively considers risks caused by label-space distribution shifts while ignoring the risks from input-space distribution shifts. On the other hand, the conventional evaluation reward detection methods for not rejecting the misclassified image in the validation dataset. However, the misclassified image can also cause risks and should be rejected. We appeal to rethink OOD detection from a human-centric perspective, that a proper detection method should reject the case that the deep model's prediction mismatches the human expectations and adopt the case that the deep model's prediction meets the human expectations. We propose a human-centric evaluation and conduct extensive experiments on 45 classifiers and 8 test datasets. We find that the simple baseline OOD detection method can achieve comparable and even better performance than the recently proposed methods, which means that the development in OOD detection in the past years may be overestimated. Additionally, our experiments demonstrate that model selection is non-trivial for OOD detection and should be considered as an integral of the proposed method, which differs from the claim in existing works that proposed methods are universal across different models.
translated by 谷歌翻译
Solving variational image segmentation problems with hidden physics is often expensive and requires different algorithms and manually tunes model parameter. The deep learning methods based on the U-Net structure have obtained outstanding performances in many different medical image segmentation tasks, but designing such networks requires a lot of parameters and training data, not always available for practical problems. In this paper, inspired by traditional multi-phase convexity Mumford-Shah variational model and full approximation scheme (FAS) solving the nonlinear systems, we propose a novel variational-model-informed network (denoted as FAS-Unet) that exploits the model and algorithm priors to extract the multi-scale features. The proposed model-informed network integrates image data and mathematical models, and implements them through learning a few convolution kernels. Based on the variational theory and FAS algorithm, we first design a feature extraction sub-network (FAS-Solution module) to solve the model-driven nonlinear systems, where a skip-connection is employed to fuse the multi-scale features. Secondly, we further design a convolution block to fuse the extracted features from the previous stage, resulting in the final segmentation possibility. Experimental results on three different medical image segmentation tasks show that the proposed FAS-Unet is very competitive with other state-of-the-art methods in qualitative, quantitative and model complexity evaluations. Moreover, it may also be possible to train specialized network architectures that automatically satisfy some of the mathematical and physical laws in other image problems for better accuracy, faster training and improved generalization.The code is available at \url{https://github.com/zhuhui100/FASUNet}.
translated by 谷歌翻译
通过自我监督的学习预先训练的大型语言模型在各种各样的任务上表现出令人印象深刻的零击功能。在这项工作中,我们介绍了Welm:一种针对中文的精心读取的预训练的语言模型,能够无缝执行不同类型的任务,以零或几次演示。 Welm通过“阅读”涵盖广泛主题的精选高质量语料库来接受10b参数的培训。我们表明,韦尔姆拥有有关各种领域和语言的广泛知识。在18个单语(中文)任务中,WELM可以大大优于现有的预训练模型,尺寸相似,并匹配高达25倍大的模型的性能。韦尔姆还表现出强大的多种语言和代码转换理解的能力,优于预先对30种语言进行预培训的现有多语言模型。此外,我们收集了人工编写的提示,并通过多次培训进行了大量的中文和微调韦尔姆的监督数据集。最终的模型可以实现对看不见的任务类型的强烈概括,并在零射门学习中优于无监督的韦尔姆。最后,我们证明韦尔姆具有解释和校准自己的决策的基本技能,这可能是未来研究的有希望的方向。我们的模型可以从https://welm.weixin.qq.com/docs/api/应用。
translated by 谷歌翻译
对抗性训练(AT)通常被认为是防御对抗性例子的最有效的方法之一,可能会在很大程度上损害标准绩效,因此对工业规模的生产和应用的有用性有限。令人惊讶的是,这种现象在自然语言处理(NLP)任务中完全相反,在该任务中甚至可以从中受益。我们注意到NLP任务中AT的优点可能来自离散和符号输入空间。为了借用NLP风格的优势,我们提出了离散的对抗训练(DAT)。 DAT利用VQGAN改革图像数据以离散类似文本的输入,即视觉单词。然后,它可以最大程度地减少这种离散图像的最大风险,并具有符号对抗扰动。我们从分布的角度进一步提供了解释,以证明DAT的有效性。作为增强视觉表示的插件技术,DAT可以在多个任务上取得重大改进,包括图像分类,对象检测和自我监督学习。尤其是,该模型通过胶带自动编码(MAE)预先训练并由我们的DAT进行微调,而没有额外的数据可以在Imagenet-C上获得31.40 MCE,并且在Stylized-Imagenet上进行了32.77%的TOP-1准确性,建立了新的状态 - 艺术。该代码将在https://github.com/alibaba/easyrobust上找到。
translated by 谷歌翻译
近年来,多视图学习迅速发展。尽管许多先前的研究都认为每个实例都出现在所有视图中,但在现实世界应用程序中很常见,从某些视图中丢失实例,从而导致多视图数据不完整。为了解决这个问题,我们提出了一个新型潜在的异质图网络(LHGN),以实现不完整的多视图学习,该学习旨在以灵活的方式尽可能充分地使用多个不完整的视图。通过学习统一的潜在代表,隐含地实现了不同观点之间一致性和互补性之间的权衡。为了探索样本与潜在表示之间的复杂关系,首次提出了邻域约束和视图约束,以构建异质图。最后,为了避免训练和测试阶段之间的任何不一致之处,基于图形学习的分类任务应用了转导学习技术。对现实世界数据集的广泛实验结果证明了我们模型对现有最新方法的有效性。
translated by 谷歌翻译
风险评分系统已被广泛地部署在许多应用程序中,这些应用程序根据用户的行为序列将风险分数分配给了。尽管许多具有复杂设计的深度学习方法已经取得了令人鼓舞的结果,但由于公平,解释性和合规性考虑,黑框的性质阻碍了他们的应用。在这些敏感情况下,基于规则的系统被认为是可靠的。但是,构建规则系统是劳动密集型的。专家需要从用户行为序列,基于统计数据的设计规则中找到信息统计信息,并为每个规则分配权重。在本文中,我们弥合了有效但黑色框模型与透明规则模型之间的差距。我们提出了一种两阶段的方法Rudi,该方法将黑框教师模型的知识提炼成基于规则的学生模型。我们设计了一种基于蒙特卡洛树搜索的统计生成方法,该方法可以在第一阶段提供一组信息统计信息。然后,通过模仿教师模型的输出,将统计数据与我们提出的神经逻辑网络组成逻辑规则。我们在三个现实世界公共数据集和一个工业数据集上评估了Rudi,以证明其有效性。
translated by 谷歌翻译
小鼠的自动社会行为分析已成为行为神经科学中越来越流行的研究领域。最近,已使用姿势信息(即关键点或骨骼的位置)来解释小鼠的社会行为。然而,很少在现有方法中研究了小鼠关键点基础的社会互动信息的有效编码和解码。特别是,由于高度变形的身体形状和模棱两可的运动模式,建模小鼠之间复杂的社交互动是一项挑战。为了处理交互建模问题,我们在这里提出了一个跨骨骼相互作用图聚合网络(CS-IGANET),以学习自由相互作用的小鼠的丰富动力学,其中使用了跨骨骼节点级交互模块(CS-NLI)建模多级相互作用(即内部,间和跨骨骼相互作用)。此外,我们设计了一种新颖的互动感知变压器(IAT),以动态学习社交行为的图形表示,并更新节点级表示,并在我们提出的互动意识到的自我注意力下的机制的指导下。最后,为了增强我们的模型的表示能力,提出了辅助自我监督的学习任务来衡量跨骨骼节点之间的相似性。标准CRMI13-SKERTON和我们的PDMB-Skeleton数据集的实验结果表明,我们所提出的模型的表现优于其他几种最先进的方法。
translated by 谷歌翻译
本文解决了高光谱(HS)图像denoising的具有挑战性的问题。与现有的基于深度学习的方法不同,通常采用复杂的网络体系结构或经验堆叠现成的模块以提高性能,我们专注于捕获HS图像的高维特性的高效提取方式。具体来说,基于理论分析,提高由展开的卷积内核形成的矩阵的排名可以促进特征多样性,我们建议分别执行1卷卷积的降级低维卷积集(Re-Convset)沿着HS图像并排的三个维度,然后通过可学习的压缩层汇总所得的空间光谱嵌入。重新汇率不仅了解HS图像的不同空间光谱特征,而且还降低了网络的参数和复杂性。然后,我们将重新汇合纳入广泛使用的U-NET体系结构中,以构建HS图像Denoisising方法。令人惊讶的是,在定量指标,视觉结果和效率方面,我们观察到这样的简洁框架在很大程度上优于最新方法。我们相信我们的工作可能会阐明基于深度学习的HS图像处理和分析。
translated by 谷歌翻译
RGB-thermal显着对象检测(RGB-T SOD)旨在定位对齐可见的和热红外图像对的共同突出对象,并准确地分割所有属于这些对象的像素。由于对热图像的照明条件不敏感,它在诸如夜间和复杂背景之类的具有挑战性的场景中很有希望。因此,RGB-T SOD的关键问题是使两种方式的功能相互补充并互相调整,因为不可避免的是,由于极端光条件和诸如极端光条件和诸如极端光明条件和热跨界。在本文中,我们提出了一个针对RGB-T SOD的新型镜子互补变压器网络(MCNET)。具体而言,我们将基于变压器的特征提取模块引入RGB和热图像的有效提取分层特征。然后,通过基于注意力的特征相互作用和基于串行的多尺度扩张卷积(SDC)特征融合模块,提出的模型实现了低级特征的互补相互作用以及深度特征的语义融合。最后,基于镜子互补结构,即使是一种模态也可以准确地提取两种方式的显着区域也是无效的。为了证明在现实世界中具有挑战性的场景下提出的模型的鲁棒性,我们基于自动驾驶域中使用的大型公共语义分段RGB-T数据集建立了一种新颖的RGB-T SOD数据集VT723。基准和VT723数据集上的昂贵实验表明,所提出的方法优于最先进的方法,包括基于CNN的方法和基于变压器的方法。该代码和数据集将在稍后在https://github.com/jxr326/swinmcnet上发布。
translated by 谷歌翻译