Machine learning in business: modern approaches and applications

This workshop is part of the CRWT workshop series held by the Centre for a Responsible Wealth Transition (CRWT) at the Telfer School of Management. This event is open to the public and is available in-person or online, however registration is required. For more information about the Centre or the Workshop Series, please contact Jonathan Li.

Machine learning has emerged as a powerful technology in today’s business world. It provides new analytics solutions to many pressing research problems in areas such as finance, operations management, healthcare, and many others. The question of how to best leverage this technology to offer new business insights has been a focal point of study in the past decade. The vast volume and variety of data available today has motivated the development of new approaches to harness the power of machine learning. This workshop is aimed at:

• introducing several business applications advanced by the latest developments of machine learning,
• introducing modern machine learning approaches, including natural language processing (NLP), large language models, deep learning, and reinforcement learning,
• providing a tutorial to those who seek to apply machine learning in their own work or research

This workshop consists of a distinguished lecture and two tutorial sessions.

1:00 pm - 2:50 pm - Distinguished lecture by Dacheng Xiu, Chicago Booth School of Business

News Text Mining: The lecture provides two examples of applying natural language processing techniques in finance. The first part focuses on sentiment analysis (classification), whereas the second part on topic modeling (clustering).

3:00 pm - 4:20 pm - Tutorial by Rafid Mahmood, NVIDIA

Modern Practices in Deep Learning: Bigger Models, Better Data: Foundation Models are an increasingly popular family of deep learning models that, by training on troves of unlabeled data, demonstrate incredible generalization to new applications. More and more deep learning applications simply adapt a general-purpose Foundation Model using custom curated data sets. This seminar will first present a tutorial of Foundation Models: how they work, and how to adapt them for our applications. In the second half of the seminar, we step back to look at the big picture of deep learning from a data-centric perspective: given the task, how do we curate, annotate, and use the appropriate data to build our model.

4:30 pm - Tutorial by Jonathan Li, Telfer School of Management

Deep reinforcement learning and business applications: This tutorial provides an overview of deep reinforcement learning (DRL) and its applications in portfolio management, financial derivatives hedging, and other dynamic decision problems. It discusses how to identify tasks that can be solved by DRL and how to implement a solution.

6:00 pm - Reception