Machine learning algorithms have grown in sophistication over the years and are increasingly deployed for real-life applications. However, when using machine learning techniques in practical settings, particularly in high-risk applications such as medicine and engineering, obtaining the failure probability of the predictive model is critical. We refer to this problem as the risk-assessment task. We focus on regression algorithms and the risk-assessment task of computing the probability of the true label lying inside an interval defined around the model's prediction. We solve the risk-assessment problem using the conformal prediction approach, which provides prediction intervals that are guaranteed to contain the true label with a given probability. Using this coverage property, we prove that our approximated failure probability is conservative in the sense that it is not lower than the true failure probability of the ML algorithm. We conduct extensive experiments to empirically study the accuracy of the proposed method for problems with and without covariate shift. Our analysis focuses on different modeling regimes, dataset sizes, and conformal prediction methodologies.
Company-as-Tribe: Company Financial Risk Assessment on Tribe-Style Graph with Hierarchical Graph Neural Networks
Institute of Computing Technology, University of Chinese Academy of Sciences(中国科学院大学计算技术研究所);Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所)
公司财务风险是普遍存在的,对上市公司进行早期的风险评估可以避免公司遭受重大损失。传统的方法主要关注企业的财务报表,缺乏对企业财务报表之间复杂关系的研究。然而,财务报表往往存在偏差和滞后,难以准确及时地识别风险。为了应对这些挑战,我们将问题重新定义为\textbf{company financial risk assessment on tribe-style graph},将每家上市公司及其股东视为一个部落,利用金融新闻建立部落间的联系。这种部落式的图表呈现出不同的模式,以区分风险公司和正常公司。然而,部落风格图中的大多数节点缺乏属性,使得难以直接采用现有的图学习方法(例如,图神经网络(GNN))。本文提出了一种新的分层图神经网络(TH-GNN),该网络分为两层,第一层通过对比学习对部落的结构模式进行编码,第二层根据部落间的关系进行信息扩散,实现了有效的风险评估.在真实公司数据集上的大量实验表明,该方法在财务风险评估方面取得了显著的改进。此外,广泛的消融研究和可视化全面显示了我们的方法的有效性。
Comparison-Based Ordinal Learning for Proactive Driving Risk Assessment
College of Automotive and Energy Engineering, Tongji University(同济大学汽车与能源工程学院);James Watt School of Engineering, University of Glasgow(格拉斯哥大学詹姆斯·瓦特工程学院)
Group-aware Contrastive Regression for Action Quality Assessment
Tsinghua University(清华大学);State Key Lab of Intelligent Technologies and Systems(智能技术与系统国家重点实验室);Beijing National Research Center for Information Science and Technology(北京信息科学与技术国家研究中心)
An evaluation of randomized machine learning methods for redundant data: Predicting short and medium-term suicide risk from administrative records and risk assessments
Dynamic Risk Assessment for Autonomous Vehicles from Spatio-Temporal Probabilistic Occupancy Heatmaps
Department of Civil and Environmental Engineering, University of California, Berkeley, CA, USA(加州大学伯克利分校土木与环境工程系);D3 Analytics Innovation, Allstate Insurance Company, Northbrook, IL, USA(Allstate保险公司D3 Analytics创新部门);Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA, USA(加州大学伯克利分校电气工程与计算机科学系)
Corneal endothelium assessment in specular microscopy images with Fuchs' dystrophy via deep regression of signed distance maps
Universidad Tecnológica de Bolívar(玻利瓦尔理工大学);University of Gothenburg(哥德堡大学);Hospital Oftalmológico Dr. Elías Santana(埃利亚斯·桑塔纳博士眼科医院);Centro Oftalmológico Virgilio Galvis(比尔希略·加尔维斯眼科中心);Fundación Oftalmológica de Santander FOSCAL(桑坦德眼科基金会FOSCAL);Universidad Autónoma de Bucaramanga UNAB(布卡拉曼加自治大学UNAB);Universidad Industrial de Santander UIS(桑坦德工业大学UIS);Universidad Politécnica de Cataluña(加泰罗尼亚理工大学)
由于存在称为guttae的暗图像区域,因此Fuchs营养不良中人角膜内皮(CE)的镜面反射显微镜评估具有挑战性。本文提出了一种基于UNet的分割方法,需要最少的后处理,并实现可靠的CE形态学评估和所有程度的富克斯营养不良的牙胶识别。我们将分割问题转换为细胞和gutta签名距离图的回归任务,而不是通常使用UNets完成的像素级分类任务。与传统的UNet分类方法相比,距离图回归方法在临床相关参数方面收敛得更快。它还产生了与手动分割的真实数据一致的形态测量参数,即平均细胞密度差异为-41.9个细胞/mm 2(95%置信区间(CI)[-306.2,222.5]),平均细胞面积的平均差异为14.8 um 2(95% CI [-41.9,71.5])。这些结果表明,CE评估的一个有前途的替代方案。