Applications of machine learning to electronic markets

Catalog description

In this 10-week course, students will learn to apply the techniques of modern machine learning (such as neural networks, reinforcement learning, generative adversarial networks, etc.) to electronic markets. Topics covered will include the fundamentals of financial electronic markets, market simulation, reinforcement learning for market-making and algorithmic execution, as well as predictive and generative modeling for financial markets. Assignments for this course will consist of a mixture of theoretical and coding exercises, and will expose students to real-life financial markets datasets. Throughout the course, students will be introduced to the latest academic and industry research papers, and will complete the course by working on a project of their choice. Course prerequisites: familiarity with optimization and statistics, and ability to code in Python. Open to Graduate Students and Senior-status Undergraduates. Others may request instructor permission to enroll.