Evaluating Active Learning Heuristics for Sequential Diagnosis
Alpen-Adria Universität Klagenfurt（阿尔卑斯-亚德里亚克拉根福大学）
Patrick Rodler, Wolfgang Schmid
备注：
This work was presented at the International Workshop on Principles of Diagnosis 2018 (DX-2018) and a version of this work was formally published as "Patrick Rodler, Wolfgang Schmid. On the impact and proper use of heuristics in test-driven ontology debugging. RuleML+RR, 2018."
Given a malfunctioning system, sequential diagnosis aims at identifying the root cause of the failure in terms of abnormally behaving system components. As initial system observations usually do not suffice to deterministically pin down just one explanation of the system's misbehavior, additional system measurements can help to differentiate between possible explanations. The goal is to restrict the space of explanations until there is only one (highly probable) explanation left. To achieve this with a minimal-cost set of measurements, various (active learning) heuristics for selecting the best next measurement have been proposed.
We report preliminary results of extensive ongoing experiments with a set of selection heuristics on real-world diagnosis cases. In particular, we try to answer questions such as "Is some heuristic always superior to all others?", "On which factors does the (relative) performance of the particular heuristics depend?" or "Under which circumstances should I use which heuristic?"
A Model-Based Active Testing Approach to Sequential Diagnosis
Delft University of Technology（代尔夫特理工大学）；University College Cork（科克大学学院）
基于模型的诊断推理往往会产生大量的诊断假设。通过考虑额外的观察（被动监测），测量额外的变量（探测）或执行额外的测试（顺序诊断/测试排序），可以减少诊断集。在本文中，我们将上述方法与自动测试模式生成（ATPG）和基于模型的诊断（MBD）的技术相结合，形成了一个名为FRACTAL（FRAamework for Active Testing AL出租）的框架。除了将系统连接到其环境的输入和输出之外，在主动测试中，我们还考虑可以向其提供测试向量序列的其他输入变量。我们解决了计算最优控制分配（定义在分形）的贪婪近似算法称为分形-G的计算困难的问题。我们比较减少剩余的最小基数诊断的FRACTAL-G的两个以上的分形算法：FRACTAL-ATPG和FRACTAL-P。FRACTAL-ATPG是基于ATPG和顺序诊断，而FRACTAL-P是基于探测，虽然不是一个积极的测试算法，提供了一个基线比较可达诊断的分形算法的数量的下限。我们经验性地评估三个分形算法的权衡进行广泛的实验ISCAS 85/74 XXX基准的组合电路。
Sequential Diagnosis Prediction with Transformer and Ontological Representation
Australian Artificial Intelligence Institute, FEIT, University of Technology Sydney（悉尼科技大学澳大利亚人工智能研究所）；School of Information Technology and Electrical Engineering, The University of Queensland（昆士兰大学信息技术与电气工程学院）
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）
ActiveMedAgent: Cost-Aware Trajectory Learning for Multimodal Medical Diagnosis
Washington University in St. Louis（圣路易斯华盛顿大学）；Icahn School of Medicine at Mount Sinai（西奈山伊坎医学院）；Arizona State University（亚利桑那州立大学）；Yale University（耶鲁大学）；Amazon（亚马逊公司）
GraD-IBD: Graph Representation Learning from Diagnosis Trajectories for Early Detection of Inflammatory Bowel Disease
Department of Surgery, Mayo Clinic, Rochester, MN, USA（外科部，梅奥诊所，罗切斯特，MN，美国）；Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, MN, USA（健康保健交付科学中心，梅奥诊所，罗切斯特，MN，美国）；Division of Hepatobiliary and Pancreas Surgery, Mayo Clinic, Rochester, MN, USA（肝胆胰外科部，梅奥诊所，罗切斯特，MN，美国）；Department of Artificial Intelligence and Informatics, Mayo Clinic, Rochester, MN, USA（人工智能与信息学部，梅奥诊所，罗切斯特，MN，美国）
HPOQuest: A Rare-Disease Diagnostic Agent Using Active Phenotype Acquisition
Technical University of Munich（慕尼黑工业大学）；Helmholtz Center Munich（亥姆霍兹慕尼黑中心）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）；Munich Data Science Institute (MDSI)（慕尼黑数据科学研究所）；Aalto University（阿尔托大学）
Discrete Equilibrium Sampling with Arbitrary Nonequilibrium Processes
D-Wave Systems Inc.（D-Wave系统公司）
我们提出了一个新的框架进行统计抽样，期望估计，分区函数近似使用\emph{arbitrary}启发式随机过程定义在离散状态空间。使用一个高度并行的结构，我们称之为\emph{sequential constraining process}，我们能够同时产生的启发式过程的状态，并准确地估计其概率，即使它们太小，实际上是通过直接计数推断。在表明理论上正确的重要性抽样和马尔可夫链蒙特卡罗是可能的顺序约束过程中，我们将其集成到一个方法称为\emph{state space sampling}，从计算机科学的抽样范围内的状态空间搜索的想法。该方法包括动态数据结构，该动态数据结构构造由受精度约束的启发式过程生成的统计的鲁棒贝叶斯模型，该精度约束为后验库尔贝克-莱布勒散度。从动态结构中采样通常会产生部分状态，这是通过递归调用启发式算法来细化结构并恢复采样来完成的。我们的各种伊辛模型的实验表明，状态空间采样，使启发式状态生成与准确的概率估计，证明了模拟退火过程的收敛到玻尔兹曼分布的运行长度增加。因此，迄今为止，前所未有的直接重要性抽样使用一般随机过程的\emph{final}（边际）分布是允许的，潜在地增加了蒙特卡罗从业者的处置的算法的范围。
