Explainable Machine Learning for Scientific Insights and Discoveries
University of Bonn（波恩大学）；University of Osnabrueck（奥斯纳布吕克大学）；University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校）；Fraunhofer Center for Machine Learning（弗劳恩霍夫机器学习中心）；Fraunhofer SCAI（弗劳恩霍夫SCAI研究所）
Ribana Roscher, Bastian Bohn, Marco F. Duarte, Jochen Garcke
Machine learning methods have been remarkably successful for a wide range of application areas in the extraction of essential information from data. An exciting and relatively recent development is the uptake of machine learning in the natural sciences, where the major goal is to obtain novel scientific insights and discoveries from observational or simulated data. A prerequisite for obtaining a scientific outcome is domain knowledge, which is needed to gain explainability, but also to enhance scientific consistency. In this article we review explainable machine learning in view of applications in the natural sciences and discuss three core elements which we identified as relevant in this context: transparency, interpretability, and explainability. With respect to these core elements, we provide a survey of recent scientific works that incorporate machine learning and the way that explainable machine learning is used in combination with domain knowledge from the application areas.
Interpretable and Explainable Machine Learning for Materials Science and Chemistry
Massachusetts Institute of Technology（麻省理工学院）；Microsoft AI for Good Research Lab（微软AI向善研究实验室）；Rutherford Appleton Laboratory（卢瑟福·阿普尔顿实验室）；University of Reading（雷丁大学）
Explainable Forecasting of Scientific Breakthroughs from Concept Network Dynamics
Geneva School of Economics and Management, University of Geneva, Geneva, Switzerland（日内瓦经济管理学院，日内瓦大学，瑞士日内瓦）；Faculty of Medicine, University of Geneva, Geneva, Switzerland（日内瓦大学医学学院，瑞士日内瓦）；TU Eindhoven, The Netherlands（埃因霍温理工大学，荷兰）；Open Quantum Institute, CERN, Geneva, Switzerland（开放量子研究所，欧洲核子研究中心，瑞士日内瓦）；armasuisse Science + Technology, Switzerland（armasuisse 科学与技术，瑞士）
Interpretable Machine Learning in Physics: A Review
University of Waterloo（滑铁卢大学）；Perimeter Institute for Theoretical Physics（周界理论物理研究所）；Homes Plus Magazine Inc.（Homes Plus杂志公司）；Santa Fe Institute（圣塔菲研究所）；ETH Zurich（苏黎世联邦理工学院）；University of Stuttgart（斯图加特大学）；Heidelberg Academy of Science and the Humanities（海德堡科学与人文学院）；Massachusetts Institute of Technology（麻省理工学院）
Unsupervised Machine Learning for Scientific Discovery: Workflow and Best Practices
Department of Neuroscience, Baylor College of Medicine（神经科学系，贝勒医学院）；Department of Applied and Computational Mathematics and Statistics, University of Notre Dame（应用与计算数学和统计学系，圣母大学）；Department of Statistics, Columbia University（统计学系，哥伦比亚大学）；Zuckerman Institute & Irving Institute, Columbia University（Zuckerman研究所与Irving研究所，哥伦比亚大学）
Classification of White Blood Cell Leukemia with Low Number of Interpretable and Explainable Features
Center for Public Health Genomics（公共卫生基因组学中心）；Department of Biomedical Engineering University of Virginia（弗吉尼亚大学生物医学工程系）；University of Virginia（弗吉尼亚大学）
Explainable AI for Curie Temperature Prediction in Magnetic Materials
Department of Data Sciences, Penn State Abington（宾夕法尼亚州立大学阿伯丁分校数据科学系）；Department of Physics, Penn State Abington（宾夕法尼亚州立大学阿伯丁分校物理系）；Department of Physics,University of Delaware（特拉华大学物理系）
Opening the Black-Box: A Systematic Review on Explainable AI in Remote Sensing
Chair of Data Science in Earth Observation, Technical University of Munich (TUM)（地球观测数据科学教授团，慕尼黑技术大学）；Munich Center for Machine Learning（慕尼黑机器学习中心）；Image Processing Laboratory (IPL), Universitat de València (UV)（图像处理实验室（IPL），瓦伦西亚大学）；University of Kaiserslautern-Landau, Germany（德国凯撒斯劳滕-兰道大学）；German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心（DFKI））；School of Applied Mathematics, Getulio Vargas Foundation, Brazil（巴西格洛里奥·瓦格斯基金会应用数学学院）；Institute for Explainable Machine Learning at Helmholtz Munich（海德堡慕尼黑可解释机器学习研究所）；Chair of Interpretable and Reliable Machine Learning, Technical University of Munich（可解释和可靠机器学习教授团，慕尼黑技术大学）
Zeroth-Order SciML: Non-intrusive Integration of Scientific Software with Deep Learning
University of Minnesota（明尼苏达大学）；Lawrence Livermore National Laboratory（劳伦斯利弗莫尔国家实验室）；Michigan State University（密歇根州立大学）；University of Michigan, Ann Arbor（密歇根大学安娜堡分校）
Decoding complexity: how machine learning is redefining scientific discovery
Alice Author Affiliation（Alice作者机构）；Tom Theory Affiliation（Tom理论机构）；Emma Experiments Affiliation（Emma实验机构）；Affiliation, Department, City, Country（机构、部门、城市、国家）；Simon Simulations Affiliation（Simon模拟机构）
Automated Molecular Concept Generation and Labeling with Large Language Models
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of California Los Angeles（加利福尼亚大学洛杉矶分校）；California Institute of Technology（加州理工学院）；Harvard University（哈佛大学）
AIGS: Generating Science from AI-Powered Automated Falsification
Department of Computer Science & Technology, Tsinghua University（清华大学计算机科学与技术系）；Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院）
Thomas Jefferson National Accelerator Facility（托马斯·杰斐逊国家加速器设施）；Massachusetts Institute of Technology（麻省理工学院）；Michigan State University（密歇根州立大学）；University of Oslo（奥斯陆大学）；The Catholic University of America（美国天主教大学）；Davidson College（戴维森学院）；William & Mary（威廉与玛丽学院）
hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devices
Fermilab（费米实验室）；Northwestern University（西北大学）；Columbia University（哥伦比亚大学）；MIT（麻省理工学院）；European Organization for Nuclear Research (CERN)（欧洲核子研究组织（CERN））
