From September 14th to November 3rd, Professor Shi Jianming, a foreign academician of the Engineering Academy of Japan and a distinguished professor of Shandong University, taught the series course titled “Fundamentals of Continuous Optimization in Machine Learning (AI)”. The course is part of the “Dacheng Spirit” Series for postgraduate capacity enhancement. More than 100 teachers and students from the School of Management, the School of Mathematics, the School of Control Science and Engineering, and the School of Mechanical Engineering attended the course.
Different from the traditional information age, machine learning, as one of the core technologies in the field of artificial intelligence, is having a significant impact on social development. Taking ChatGPT as an example, the course introduced the current problem-solving capabilities of AI and its applications in the field of management, with continuous optimization methods essential for building machine learning (AI) algorithms as the core.
Stressing that optimization is at the core of AI and taking a support vector machine as an example, Professor Shi introduced the three elements of machine learning models and the derivation of the basic quantities of a support vector machine. He analyzed several optimization methods commonly used in the machine learning, including Local Optima and Global Optima of the objective function within a region, as well as Steepest Descent Method, Least Squares, Gradient Descent, and other iterative algorithms. He also focused on the neural network algorithm and the detailed steps of using EXCEL for calculation.
During the question and answer session, Professor Shi discussed with students about various continuous optimization methods in AI algorithms, the setting and premise of mathematical analysis problems, as well as the cutting-edge technologies and development trends in the field of AI algorithms. In the assessment stage of the course, Professor Shi guided students to complete the programming with neural network algorithms.
The series course “Fundamentals of Continuous Optimization in Machine Learning (AI)” has a total of 16 class hours with in-depth and interesting content, which broadens the academic horizons of postgraduates, and improves their academic and innovation ability.
