MATDAT18: Materials and Data Science Hackathon

材料研究部，信息和智能系统部以及数学科学部为该奖项提供资金。它支持组织“数据黑客”，其中每个由材料研究人员和数据科学家组成的跨学科团队将共同努力，应用先进的数据科学方法来解决材料研究本质上跨学科领域的重要挑战性问题。本次黑客活动的目标是激发新的合作，通过数据科学的领先方法以新颖的方式解决材料科学中重要的、具有挑战性的问题。材料数据和可用计算能力的指数级增长使生成和分析大量材料数据成为可能。材料项目等倡议已经创建了可公开访问的数据库，其中包含数万种材料的结构和性质。除了这些通用资源，个别研究小组正在为更具体的材料研究问题生成大型数据集。材料科学与工程领域的主要挑战之一是确定如何在设计和发现具有所需特性的新材料的过程中最好地利用这些丰富的材料数据，并加速其在现有技术中的部署和创新新技术。尽管近年来数据分析和机器学习在材料科学中的应用取得了相当大的进展，但仍然存在一个根本问题，即大多数材料科学和工程专家不是数据科学专家，数据科学家也不是材料研究专家。这项“数据黑客”活动旨在将材料研究界和数据科学界聚集在一起，共同解决以数据为中心的方法可能取得进展的重大问题，并促进社区之间的合作。这种“数据黑客”将从根本上跨学科的建设。代表数据和材料科学的研究人员小组将在实质性的材料问题上面对面工作几天。希望材料研究人员能够接触到尖端的统计和机器学习技术，数据科学家也有动力开发新方法来分析材料研究人员产生的新数据流。参与者申请过程还将强调纳入初级研究人员和代表性不足的群体。

The Division of Materials Research, the Division of Information and Intelligent Systems, and the Division of Mathematical Sciences contribute funds to this award. It supports the organization of a "data hackathon" in which interdisciplinary teams, each composed of both materials researchers and data scientists, will work together to apply advanced data science methods to address important, challenging problems in the intrinsically interdisciplinary field of materials research. The goal of this hackathon is to spark new collaborations in which important, challenging problems in materials science are addressed in novel ways by leading methods in data science.The exponential increase in materials data and in available computing power has made it possible to generate and analyze large amounts of materials data. Initiatives such as the Materials Project have created publicly accessible databases containing the structure and properties of tens of thousands of materials. In addition to these general-purpose resources, individual research groups are generating large data sets for more specific materials research problems. One of the leading challenges in materials science and engineering is determining how to best make use of this abundance of materials data in the process to design and discover new materials with desired properties and accelerate their deployment in existing technologies and innovate new technologies. Despite the considerable progress that has been made in the application of data analytics and machine learning to materials science in recent years, there is still a fundamental problem in that most experts in materials science and engineering are not experts in data science, and data scientists are not experts in materials research. This "data hackathon" activity is aimed to bring the materials research community and the data science community together to attack significant problems where a data-centric approach may be able to make progress and to stimulate collaboration among the communities. This "data hackathon" will be fundamentally interdisciplinary by construction. Small groups of researchers representing both data and materials sciences will work face-to-face for several days on substantial materials problems. It is hoped that materials researchers are exposed to cutting-edge statistics and machine-learning techniques, and data scientist are motivated to develop new methods to analyze the novel data streams produced by materials researchers. The participant application process will also emphasize inclusion of junior researchers and under-represented groups.