������Ϣ�W��Փ(��):�C���W�����A(Ӣ�İ�) ���x

Chapter 1 Introduction Materials informatics is an emerging and rapidly developing field�� particularly after the launch of Materials Genome Initiative (MGI) in 2011. Ramakrishna et al. (2019) defined materials informatics as that ��the materials informatics employs techniques�� tools�� and theories drawn from the emerging fields such as data science�� internet�� computer science and engineering�� and digital technologies to the materials science and engineering to accelerate materials�� products and manufacturing innovations��. Agrawal and Choudhary (2016) proposed that materials informatics has become the ��fourth paradigm�� in the research and development of materials. Following the guidance of MGI�� many free-access and open-to-public material databases have been built up�� and materials data are exponentially growing thanks to the development of high-throughput experiments and high-throughput calculation techniques�� as well as data sharing among the materials communities. A list of available materials databases can be found in the review paper of Ramakrishna et al. (2019). Materials informatics integrates materials science and engineering and the sciences and technologies from artificial intelligence (AI)�� database�� and machine learning (ML) to accelerate the innovation in the whole materials development continuum from discovery�� development�� property optimization�� system design and integration�� manufacturing to deployment�� and to speed up the process from data to knowledge in order to understand and master the relationship between material micro-structures and macro-properties based on material composition�� processing�� and performance. Materials informatics adds the novel tools of AI and ML into the toolbox of materials science and engineering�� which will definitely strengthen and enhance the power of the methodologies in materials research and development. Materials informatics utilizes AI and ML to analyze a large ensemble of materials data from experiments�� computations�� manufactures�� industries�� and daily life�� etc.�� efficiently and cost-effectively�� and to deliver materials knowledge and technology in user-friendly ways to designers�� scientists�� engineers�� and manufacturers of materials and products. ML gives computers the ability to learn from data and make predictions based on data (Samuel 1967). Mitchell (1997) defined ML as follows: ��A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P�� if its performance at tasks in T�� as measured by P�� improves with experience E.�� In materials informatics�� ML algorithms learn from existing material data with some properties or/and performance of interest�� in order to improve targeted properties and performance of existing materials or/and to design and discover new materials with desired properties and performance. Since material data are usually small�� adaptive design with feedback from experiments or/and computations has been proposed to enhance the ML ability. Figure 1.1 shows the adaptive learning and design in materials informatics. Initially successful and failed data�� with material properties and performance of interest and material composition�� processing�� testing conditions�� or/and service environments�� etc. of adjustable parameters�� are generated from experiments�� calculations�� or/and manufacturing. A dataset is usually built up by one��s own produced data or/and by collecting data from various sources. The material properties and performance of interest are usually called output variables�� and the parameters of material composition�� processing�� or/and testing conditions are normally called input variables. In addition�� many molecular�� atomic and electronic parameters�� thermodynamic properties�� kinetic parameters�� crystalline and amorphous information�� etc. are often employed as input or/and output variables. ML is conducted on an initial dataset. The learning results will be evaluated and interpreted automatically or/and by domain experts�� who are authorities in a particular area or topic�� and clearly�� domain experts in materials informatics are materials experts and scientists. The adaptive learning process is an iterative and interactive loop. The learning results are adopted to design and guide next experiments�� calculations�� or/and manufacture�� and the new results will validate the ML prediction and simultaneously be fed back to the dataset for the next cycle of iterations in the adaptive learning loop�� which means that the dataset will grow after each iteration. Obviously�� the next round of adaptive learning may give a different result because of more data added�� which provides modified guidance on experiments�� calculations�� or/and manufacture. This means that each step in the adaptive design loop will refine the inputs and outputs more or less during the cycle of iterations�� and the iterations will go continuously until the designed goal of material properties is reached. Lookman