Course Description

This advanced course delves into the cutting-edge techniques of machine learning with a specific focus on their applications in finance. Expanding upon foundational statistical learning methods, the course covers deep learning architectures such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), including long short-term memory (LSTM) networks. Through a combination of theoretical lectures, coding assignments, and projects using real-world financial data, students will gain expertise in applying machine learning algorithms to challenges in asset pricing, risk management, and portfolio optimization. This course will also address recent advancements in LLMs and their integration into agent workflows and decision-making processes in the finance industry.

Prerequisites: Students are expected to have prior knowledge of machine learning and statistical learning as taught in a course such as FA 590 Statistical Learning in Finance. A quantitative background is required, including knowledge of probability, statistics, and calculus. Prior programming experience in at least one programming language (e.g., Python or R) is mandatory.

Student Learning Outcomes

Upon successful completion of this course, students will be able to:

• Develop and implement machine learning models, including neural networks and deep learning architectures, for solving complex financial problems.
• Integrate and fine-tune large language models (LLMs) for financial applications, including textual analysis and decision-making workflows
• Evaluate the performance of various machine learning algorithms in finance, selecting the most suitable models based on accuracy, efficiency, and scalability
• Understand the ethical and practical implications of applying machine learning in finance,particularly regarding data privacy, interpretability, and regulatory compliance.

Textbook

• Ian Goodfellow, Yoshua Bengio, and Aaron Courville. Deep Learning. MIT Press. 2016.1

Generative AI Technologies

ChatGPT and other generative AI (GenAI) tools are allowed and encouraged in this course. However, you should be aware that the material generated by GenAI may be inaccurate, incomplete, or otherwise problematic. You need to disclose in each of the assignment and the final project on how they have used GenAI, reflect and summarize good use cases of GenAI, if any. Using GenAI without disclosing it is considered plagiarism and will be dealt with under relevant Stevens policies.