Active Diagnosis via AUC Maximization: An Efficient Approach for Multiple Fault Identification in Large Scale, Noisy Networks
University of Michigan(密歇根大学);University of Texas Medical Branch(德克萨斯大学医学部);Institute for Translational Sciences(转化科学研究所);Electrical Engineering and Computer Science(电气工程与计算机科学学院)
Gowtham Bellala, Jason Stanley, Clayton Scott, Suresh K. Bhavnani
The problem of active diagnosis arises in several applications such as disease diagnosis, and fault diagnosis in computer networks, where the goal is to rapidly identify the binary states of a set of objects (e.g., faulty or working) by sequentially selecting, and observing, (noisy) responses to binary valued queries. Current algorithms in this area rely on loopy belief propagation for active query selection. These algorithms have an exponential time complexity, making them slow and even intractable in large networks. We propose a rank-based greedy algorithm that sequentially chooses queries such that the area under the ROC curve of the rank-based output is maximized. The AUC criterion allows us to make a simplifying assumption that significantly reduces the complexity of active query selection (from exponential to near quadratic), with little or no compromise on the performance quality.
MBZUAI(穆罕默德·本·扎耶德人工智能大学);Nanjing University of Information Science & Technology(南京信息工程大学);Harbin Institute of Technology(哈尔滨工业大学);University of Pittsburgh(匹兹堡大学)
Adaptive Subgradient Methods for Online AUC Maximization
The University of Chicago(芝加哥大学);Singapore Management University(新加坡管理大学);Institute for Infocomm Research, A*STAR(新加坡科技研究局资讯通信研究院);Nanyang Technological University(南洋理工大学)
A Robust AUC Maximization Framework with Simultaneous Outlier Detection and Feature Selection for Positive-Unlabeled Classification
University of Rochester(罗切斯特大学);Baylor College of Medicine(贝勒医学院);University of Texas Health Science Center at Houston(休斯顿德克萨斯大学健康科学中心);University of Washington(华盛顿大学)
Communication-Efficient Federated AUC Maximization with Cyclic Client Participation
Department of Computer Science, Northern Illinois University(北伊利诺伊大学计算机科学系);Department of Computer Science, University of North Carolina at Charlotte(北卡罗来纳大学夏洛特分校计算机科学系);Center for Research in Computer Vision, University of Central Florida(佛罗里达中央大学计算机视觉研究中心)
Federated Deep AUC Maximization for Heterogeneous Data with a Constant Communication Complexity
The University of Iowa(爱荷华大学);DAMO Academy, Alibaba Group(阿里巴巴集团达摩院);State University of New York at Albany(纽约州立大学奥尔巴尼分校)
深度AUC(area under the ROC curve)最大化(DAM)方法由于其在不平衡数据分类中的巨大潜力而受到广泛关注。然而,联邦深度AUC最大化(FDAM)的研究仍然有限。与关注可分解最小化目标的标准联邦学习(FL)方法相比,FDAM由于其最小化目标在单个示例上不可分解而更加复杂。在本文中,我们提出了改进的FDAM算法的异构数据,通过解决流行的非凸强凹的min-max制定的DAM在分布式的方式,这也可以应用到一类非凸强凹min-max问题。本文的一个引人注目的结果是,所提出的算法的通信复杂性是一个常数,独立于机器的数量,也独立于精度水平,这改善了一个现有的结果的数量级。实验证明了我们的FDAM算法的有效性基准数据集上,并从不同组织的医学胸部X射线图像。我们的实验表明,使用来自多个医院的数据的FDAM的性能可以提高来自单个医院的测试数据的AUC分数,用于基于胸片检测危及生命的疾病。所提出的方法在我们的开源库LibAUC(www.libauc.org)中实现,其github地址为https://github.com/Optimization-AI/ICML2021_FedDeepAUC_CODASCA。
A Game-Theoretic Analysis of the Empirical Revenue Maximization Algorithm with Endogenous Sampling
Peking University(北京大学);Technion - Israel Institute of Technology(以色列理工学院);Hong Kong University of Science and Technology(香港科技大学);Alibaba Group(阿里巴巴集团);Shanghai Jiao Tong University(上海交通大学)
经验收益最大化(ERM)是拍卖设计中最重要的价格学习算法之一:文献表明,它可以在重复拍卖和统一价格拍卖中学习到收益最大化拍卖者的近似最优保留价。然而,在这些应用程序中,提供输入的机构有动机操纵输入,以降低输出的价格。我们概括了Lavi et al(2019)提出的激励意识度量的定义,以量化由于$N$输入样本中的$m\ge 1$的变化而导致的ERM输出价格的降低,并提供了该度量到零的特定收敛率,因为对于不同类型的输入分布,$N$趋于无穷大。通过采用这一措施,我们构建了一个有效的,近似激励兼容的,和收入最优的学习算法使用ERM在重复拍卖对非近视投标人,并显示近似组激励兼容统一价格拍卖。
Provable Multi-instance Deep AUC Maximization with Stochastic Pooling
University of Iowa(爱荷华大学);Texas A&M University(德克萨斯农工大学);Beijing Institute of Ophthalmology(北京眼科研究所);Beijing Tongren Hospital(北京同仁医院);Capital Medical University(首都医科大学)
BEAT-Net: Injecting Biomimetic Spatio-Temporal Priors for Interpretable ECG Diagnosis
School of Information Technology(信息科技学院);Monash University Malaysia(墨尔本大学马来西亚分校);Faculty of Biomedical Engineering(生物医学工程学院);Shenzhen University of Advanced Technology(深圳先进技术大学)
Calibration Matters: Tackling Maximization Bias in Large-scale Advertising Recommendation Systems
Carnegie Mellon University(卡内基梅隆大学);University of Chicago Booth School of Business(芝加哥大学布斯商学院);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学)
Partial AUC optimization based deep speaker embeddings with class-center learning for text-independent speaker verification
Northwestern Polytechnical University(西北工业大学)
基于深度嵌入的文本无关说话人确认在许多具有挑战性的场景中表现出优于传统方法的性能。其损失函数一般可分为两类,即,核查和识别。验证损失函数与说话人验证的流水线相匹配,但其实现比较困难。因此,大多数最先进的深度嵌入方法使用具有softmax输出单元或其变体的识别损失函数。在本文中,我们提出了一个验证损失函数,命名为最大化的接收机操作特征(ROC)曲线下的部分面积(pAUC),基于深度嵌入文本无关的说话人验证。我们还提出了一种基于类中心的训练试验构建方法,以提高训练效率,这是关键的建议损失函数是可比的识别损失的性能。在Speaker in the Wild(SITW)和NIST SRE 2016数据集上的实验表明,所提出的pAUC损失函数与最先进的识别损失函数具有很强的竞争力。