Speaker: Chong Woon Han
Title: Building a Trading Algorithm with Binary Classification
Abstract: My Capstone explores the effectiveness of using machine learning on trading algorithms. I will introduce how I built a machine learning classifier to predict whether a stock in the US Equities market will rise or fall. I will then show whether applying the classifier to a trade algorithm on backtest data can lead to better returns than when a simple momentum-based approach or a random classifier is used instead.
Speaker: Ong Kaijing Gin
Title: Applications of Machine Learning to Extremal Graph Theory
Abstract: Ramsey theory is a branch of mathematics studying the mathematical principle that no matter how we partition the objects of a “large” structure into a “few” classes, one of these classes contains a “large” subsystem (Bollobás, 1998). A key result in Ramsey theory states one will find monochromatic cliques in every red-blue coloring of a sufficiently large complete graph. How large this complete graph must be is known for monochromatic cliques with 3 and 4 vertices. However, this problem has yet to be solved for larger cliques. My project explores if we can train a computer to solve such problems in Ramsey theory.