What Can DP Do too? | From Principle Deduction to Practical Application: Professor Shan Bin's 2025 New Course Material Big Data and Machine Learning Now Available!
In an era of increasing intersection between materials science and artificial intelligence, machine learning technologies are gradually becoming essential tools for materials research. To help students and researchers explore new approaches for materials design, discovery, and optimization from an artificial intelligence perspective using machine learning methods, Professor Shan Bin from Huazhong University of Science and Technology will offer the course Material Big Data and Machine Learning offline on May 12, 2025. As always, course content will be updated online simultaneously to facilitate online learning for interested learners worldwide.
Professor Shan Bin has made outstanding research achievements and possesses extensive teaching experience in the interdisciplinary field of materials science and machine learning. His Computational Materials Science course at Huazhong University of Science and Technology has been meticulously refined over more than a decade, with content presented in an accessible yet profound manner, making it a highly influential open course in the field of computational materials science. Since its launch on the Bohrium platform, the course has attracted over 7,000 learners. Additionally, the supporting course for Professor Shan Bin’s book Computational Materials Science: From Algorithm Principles to Code Implementation is also available on the Bohrium platform, covering additional details of the computational materials science curriculum.
How Does the New Course Differ from Computational Materials Science?
Compared to the annually updated Computational Materials Science, one of the key highlights of this course is its tight focus on the practical needs of materials science, deeply integrating machine learning technologies with materials science problems. Through rigorous algorithm instruction and rich case studies, the course focuses on real-world challenges in the materials field, allowing learners to intuitively experience the applications of machine learning in material property prediction, material structure analysis, material design optimization, and more. It aims to cultivate learners’ comprehensive capabilities to solve practical materials problems using machine learning.
What Does the Course Cover?
The application of machine learning in materials science is complex and multifaceted. This course spans core content from the fundamental theories of machine learning to its specific applications in materials science. Starting with the concepts and principles of machine learning, it explains the theoretical foundations of linear regression and its applications in materials science. Core modules cover feature engineering and model tuning, support vector machines, clustering algorithms, and other key topics. The course also incorporates cutting-edge artificial neural network technologies, combined with materials science case studies, enabling learners to proficiently use neural networks to build complex machine learning models and address intricate problems in materials science.
Course Syllabus:
1.Introduction to Machine Learning
2.Data Analysis Essentials (Pandas, NumPy, Matplotlib)
3.Linear Regression and Least Squares
4.Linear Regression Case Studies in Materials Science
5.Regularization Topics
6.Theoretical Foundations of Logistic Regression
7.(Optional) KNN Classification Algorithm
8.Logistic Regression Case Studies in Materials Science
9.Feature Engineering and Model Optimization
10.Support Vector Machines
11.SVM Case Studies and Kernel Methods
12.Clustering Algorithms
13.Decision Tree Methods
14.Decision Tree Case Studies in Materials Science
15.Ensemble Learning Methods
16.Foundations of Neural Networks
17.Neural Network Case Studies in Materials Science
Why Take This Course?
Beyond its systematic and comprehensive content, the course maintains a practice-oriented teaching model. Professor Shan Bin emphasizes the importance of hands-on experience in mastering machine learning. Through Bohrium Notebook—a powerful practical tool—substantial hands-on coding sessions are integrated into the curriculum. Learners can run Notebooks online using CPU or GPU cores in the Bohrium environment, reproduce and practice course cases directly, and freely save, create, and share Notebook content to build personalized "learning portfolios." This approach deepens understanding of theoretical knowledge through practical application, ensuring learners master every detail from theory to implementation and truly integrate learning with practice.
How to Enroll?
Click the link below to access the course:
Course Link
https://j1q.cn/eWTv3Fd2
Upon enrollment, a QR code page for the [2025 Material Big Data and Machine Learning Classroom Group] will pop up. Scan the code to join the group.
(If you can't join via the QR code, please add the Deep Potential administrator's WeChat ID: deeppotential. The administrator will manually invite you to the group.)
FAQs
1.Is the course free?
Continuing the tradition of open sharing, this course is offered free of charge.
2.Who should take this course?
This course is designed for learners interested in applying machine learning to materials science, including undergraduate and graduate students in materials science, as well as researchers and professionals engaged in materials R&D and applications.
3.What is the course update schedule?
As the course is taught offline first, online content will be organized and released after each offline session. New course materials (videos, lecture notes, Notebooks, etc.) will be uploaded every Monday and Wednesday at 19:00. Please allow for minor delays in content synchronization.
4.Can I catch up if I start late?
Yes. The online course uses pre-recorded videos, and all content (videos, courseware, code examples) will be available indefinitely. You can join and start learning at any time.
We wish all learners a rewarding experience!
New Book Announcement
Machine Learning-Assisted Materials Design, co-authored by DeepModeling, Huazhong University of Science and Technology, and Northwestern Polytechnical University, and published by Tsinghua University Press, will be released soon. The book systematically introduces programming tools and artificial intelligence algorithms applicable to materials science and explores the deep integration of machine learning with materials research. Stay tuned!