Artificial Intelligence: Scientific Data Analysis and Machine Learning Applications

Course 12: Artificial Intelligence: Scientific Data Analysis and Machine Learning Applications

I. Course Description

Just as the 19th century invented machines that could multiply the mechanical capabilities of humans, the last half-century has witnessed the emergence of machines and technologies that have doubled our ability to collect, analyze, and understand data. Over the past decade, the importance and potential of discovering computational knowledge in large datasets has grown further exponentially, driven by faster and cheaper computer and data storage, the development of accessible languages / frameworks (such as Python), and the explosive growth in machine learning and artificial intelligence.

In the course, students will understand the goals and basic principles of data analysis, explore the basic ideas and principles of machine learning, the connection of classification and regression tasks, and examples of supervised and unsupervised learning. The professor will introduce students to Python and some of the key data analysis libraries to understand and master the different software used. It is then discussed in the course that techniques will be applied to realistic examples using large datasets from different scientific fields.

II. Professor Introduction

Gunther Roland – Tentenured professor at MIT

Professor Gunther Roland received his PhD from the Kernphysik Institute in Frankfurt, joined the Heavy Ion Group from the MIT Department of Physics in September 2000 and served as a scientific assistant to the group. The professor now serves as the joint leader of seven research groups, including the MIT Heavy Ion Research Group.

He is also Chairman of the CMS Heavy Ion Publishing Committee; Head of the Quantum Physics Experiment Program sPHENIX; Member, member of the Editorial Board of Annual Rev. Nucl. Part. Phys.

III. Syllabus

1. An Introduction to Python Programming
2. Data analysis and basic statistics
3. Data visualization and large data sets
4. Introduction to multivariate analysis
5. Introduction to machine learning
6. 6.Scikit-learn
7. supervised learning and classical models
8. Unsupervised learning and classical models
9. Reinforcement learning: Markov decision-making process, Q learning
10. Deep learning: CNN, RNN, and GAN