Source-free domain adaptation (SFDA) aims to transfer knowledge learned from a source domain to an unlabeled target domain, where the source data is unavailable during adaptation. Existing approaches for SFDA focus on self-training usually including well-established entropy minimization techniques. One of the main challenges in SFDA is to reduce accumulation of errors caused by domain misalignment. A recent strategy successfully managed to reduce error accumulation by pseudo-labeling the target samples based on class-wise prototypes (centroids) generated by their clustering in the representation space. However, this strategy also creates cases for which the cross-entropy of a pseudo-label and the minimum entropy have a conflict in their objectives. We call this conflict the centroid-hypothesis conflict. We propose to reconcile this conflict by aligning the entropy minimization objective with that of the pseudo labels' cross entropy. We demonstrate the effectiveness of aligning the two loss objectives on three domain adaptation datasets. In addition, we provide state-of-the-art results using up-to-date architectures also showing the consistency of our method across these architectures.
translated by 谷歌翻译
神经网络量化能够在边缘设备上部署模型。对其硬件效率的基本要求是平衡器是硬件友好的:均匀,对称,以及两个阈值的功率。据我们所知,目前的训练后量化方法不同时支持所有这些约束。在这项工作中,我们引入了硬件友好的训练量化(HPTQ)框架,通过协同组合几种已知的量化方法来解决这个问题。我们对四个任务进行了大规模的研究:在各种网络架构上进行分类,对象检测,语义分割和姿势估计。我们广泛的实验表明,可以在硬件友好的限制下获得竞争结果。
translated by 谷歌翻译
Pairwise compatibility measure (CM) is a key component in solving the jigsaw puzzle problem (JPP) and many of its recently proposed variants. With the rapid rise of deep neural networks (DNNs), a trade-off between performance (i.e., accuracy) and computational efficiency has become a very significant issue. Whereas an end-to-end DNN-based CM model exhibits high performance, it becomes virtually infeasible on very large puzzles, due to its highly intensive computation. On the other hand, exploiting the concept of embeddings to alleviate significantly the computational efficiency, has resulted in degraded performance, according to recent studies. This paper derives an advanced CM model (based on modified embeddings and a new loss function, called hard batch triplet loss) for closing the above gap between speed and accuracy; namely a CM model that achieves SOTA results in terms of performance and efficiency combined. We evaluated our newly derived CM on three commonly used datasets, and obtained a reconstruction improvement of 5.8% and 19.5% for so-called Type-1 and Type-2 problem variants, respectively, compared to best known results due to previous CMs.
translated by 谷歌翻译
给定仿射转换$ t $,我们定义其Fisher失真$ dist_f(t)$。我们表明,Fisher失真具有Riemannian度量结构,并提供了一种用于查找平均变形转换的算法 - 即 - 对于给定的$ \ {t_ {i} \} _ {i = 1}^n $的仿射转换,,,找到一个仿射转换$ t $最小化整体失真$ \ sum_ {i = 1}^ndist_f^{2}(t^{ - 1} t_ {i})。$平均变形转换在某些字段中可以很有用 - 特别是,我们将其应用于渲染仿射全景。
translated by 谷歌翻译
与人类类似,动物的面部表情与情绪状态紧密相关。但是,与人类领域相反,动物面部表情对情绪状态的自动识别是没有充满反应的,这主要是由于数据收集和建立地面真相的困难,涉及非语言用户的情绪状态。我们将最近的深度学习技术应用于在受控的实验环境中收集的数据集上对狗的挫败进行分类和(负面)的挫败感。我们探索在此任务的不同监督下不同骨干(例如,重新连接,VIT)的适用性,并发现自我监督的预定的VIT(DINO-VIT)的特征优于其他替代方案。据我们所知,这项工作是第一个解决对受控实验中获得的数据自动分类的任务。
translated by 谷歌翻译
联合学习已被提议作为隐私的机器学习框架,该框架使多个客户能够在不共享原始数据的情况下进行协作。但是,在此框架中,设计并不能保证客户隐私保护。先前的工作表明,联邦学习中的梯度共享策略可能容易受到数据重建攻击的影响。但是,实际上,考虑到高沟通成本或由于增强隐私要求,客户可能不会传输原始梯度。实证研究表明,梯度混淆,包括通过梯度噪声注入和通过梯度压缩的无意化混淆的意图混淆,可以提供更多的隐私保护,以防止重建攻击。在这项工作中,我们提出了一个针对联合学习中图像分类任务的新数据重建攻击框架。我们表明,通常采用的梯度后处理程序,例如梯度量化,梯度稀疏和梯度扰动,可能会在联合学习中具有错误的安全感。与先前的研究相反,我们认为不应将隐私增强视为梯度压缩的副产品。此外,我们在提出的框架下设计了一种新方法,以在语义层面重建图像。我们量化语义隐私泄漏,并根据图像相似性分数进行比较。我们的比较挑战了文献中图像数据泄漏评估方案。结果强调了在现有联合学习算法中重新审视和重新设计对客户数据的隐私保护机制的重要性。
translated by 谷歌翻译
This paper introduces the novel CNN-based encoder Twin Embedding Network (TEN), for the jigsaw puzzle problem (JPP), which represents a puzzle piece with respect to its boundary in a latent embedding space. Combining this latent representation with a simple distance measure, we demonstrate improved accuracy levels of our newly proposed pairwise compatibility measure (CM), compared to that of various classical methods, for degraded puzzles with eroded tile boundaries. We focus on this problem instance for our case study, as it serves as an appropriate testbed for real-world scenarios. Specifically, we demonstrated an improvement of up to 8.5% and 16.8% in reconstruction accuracy, for so-called Type-1 and Type-2 problem variants, respectively. Furthermore, we also demonstrated that TEN is faster by a few orders of magnitude, on average, than a typical deep neural network (NN) model, i.e., it is as fast as the classical methods. In this regard, the paper makes a significant first attempt at bridging the gap between the relatively low accuracy (of classical methods and the intensive computational complexity (of NN models), for practical, real-world puzzle-like problems.
translated by 谷歌翻译
后空飞行是一种水生昆虫,能够在水下调节其浮力。它的腹部被血红蛋白细胞覆盖,用于啮合和释放氧气,可逆地。进入水后,飞捕口在其腹部的超疏水毛状结构中的气泡进行呼吸。然而,这种泡沫可以通过来自腹部血红蛋白细胞的调节氧气流动来改变其体积。通过这种方式,它可以达到中性浮力而无需进一步的能量消耗。在这项研究中,我们开发了一种小,厘米的刻度,通过受控成核和释放微泡的自动浮力调节来发展一小厘米。气泡通过电解,直接在板载电极上直接生长,通过低电压调节。我们使用3D打印来引入三维气泡诱捕的蜂窝结构,以创造一个稳定的外部气体储层。为了减少浮力力,气泡通过线性机械振动释放,从机器人的身体分离。通过压力传感和比例整体衍生控制回路机构,机器人自动调节其浮力,以在几秒钟内水下达到中性浮动。这种机制可以促进更换传统和物理上更大的浮力调节系统,如活塞和加压罐,并能够实现自主水下车辆的小型化。
translated by 谷歌翻译
谷歌的运营洪水预测系统是制定的,为机构和公众提供准确的实时洪水警告,重点是河流洪水在大型潮流的河流中。它在2018年开始运作,自从地理位置扩展以来。该预测系统由四个子系统组成:数据验证,阶段预测,淹没建模和警报分配。机器学习用于两个子系统。阶段预测采用长短期内存(LSTM)网络和线性模型进行建模。使用阈值和歧管模型计算洪水淹没,前者计算淹没程度,后者计算淹没程度和深度。本文首次提供的歧管模型提供了一种机器学习替代洪水淹没的液压建模。在评估历史数据时,所有型号都可以实现可操作使用的足够高的度量指标。 LSTM表现出比线性模型更高的技能,而阈值和歧管模型达到了类似的性能度量,以便在淹没程度上进行建模。在2021年的季风季节期间,洪水预警系统在印度和孟加拉国运营,覆盖河流的洪水区,总面积287,000平方公里,拥有350多万人。超过100米的洪水警报被发送给受影响的人口,相关当局以及紧急组织。系统上的当前和未来的工作包括将覆盖范围扩展到额外的洪水易发位置,以及提高建模能力和准确性。
translated by 谷歌翻译
联合学习(FL)是一种保护隐私的范式,其中多个参与者共同解决机器学习问题而无需共享原始数据。与传统的分布式学习不同,FL的独特特征是统计异质性,即,跨参与者的数据分布彼此不同。同时,神经网络解释的最新进展已广泛使用神经切线核(NTK)进行收敛分析。在本文中,我们提出了一个新颖的FL范式,该范式由NTK框架赋予了能力。该范式通过传输比常规FL范式更具表现力的更新数据来解决统计异质性的挑战。具体而言,通过样本的雅各布矩阵,而不是模型的权重/梯度,由参与者上传。然后,服务器构建了经验内核矩阵,以更新全局模型,而无需明确执行梯度下降。我们进一步开发了一种具有提高沟通效率和增强隐私性的变体。数值结果表明,与联邦平均相比,所提出的范式可以达到相同的精度,同时将通信弹的数量减少数量级。
translated by 谷歌翻译