AI for Financial Services

Greg Mori, Borealis AI and Simon Fraser University, Canada

Challenges and Opportunities in Machine Learning for Financial Services

Financial services are at the core of our economy.  Opportunities for machine learning abound in this space, from capital markets to insurance services to wealth management to lending to tools that assist clients in managing their money.  Modern machine learning methods have transformed industries, yet particular challenges exist in realizing the full potential of machine learning in financial services.  In this talk I will describe some of the main technical challenges for machine learning in financial services, including explainability, data imbalance, partial observations, distribution shift, and self-supervised learning in low-signal settings.

Bo An, Nanyang Technological University, Singapore

Recent Progress on Reinforcement Learning for Quantitative Trading

In the last decade, we have witnessed a significant development of AI-powered quantitative trading (QT), due to its instant and accurate order execution, and capability of analyzing and processing large amount of data related to the financial market. Traditional AI-powered QT methods discover trading opportunities based on either heuristic rules or financial prediction. However, due to the high volatility and noisy nature of financial market, their performance is not stable and highly reply on the market condition. Recently, reinforcement learning (RL) becomes an appealing approach for QT tasks owing to its stellar performance on solving complex decision-making problems. This talk will discuss some recent research progress in RL for QT and future directions.

J.B. Heaton, One Hat Research LLC, USA

Market Evidence and Feature Engineering for Predictive Finance: Bankruptcy/Default Prediction and Insolvency Detection

Artificial intelligence and complex financial models play an increasingly prominent role in investment research, including bankruptcy and default prediction and insolvency detection. Existing efforts often overlook the powerful features available in efficient or nearly-efficient markets that provide direct predictions and measurements of target variables of interest. I explore four such powerful features: stock returns that predict bankruptcy better than existing models, lower bounds on probability of default implied by bond prices, and insolvency detection using both bond prices compared to lower bounds on those prices for a balance-sheet solvent firm and spreads between secured and unsecured debt prices that evidence the value of lien in a likely bankruptcy. The results remind us that market evidence is always likely to exceed the performance of economists' models where it bears directly on a predictive problem like the wipeout of equity value, debt default, and the impact of bankruptcy on unsecured creditors.

Pamela Yoon, Royal Bank of Canada

What is Wealth Management and How Can AI Play a Role in this Space?

In this talk, she will describe her role in wealth management, her team and a few current client scenarios. She will also address the reasons for needing an advisor, costs of the advisor and is it worth it, different life stages of clients, and how we help;

- Starting out/accumulating wealth

- Starting a family

- Clients heading into their career peak, busy with career priorities

- Tax issues (Canada is a high tax regime), understanding tax strategies is very important

- End of life planning

- Risk management – if something unexpected (and bad) happens, how does that shatter dreams.

Yongjae Lee, Ulsan National Institute of Science and Technology, South Korea

Investment Recommendations for Individual Investors

Investment behaviors of individual investors vary as widely as their personalities. To develop personalized investment services, three key factors should be considered: 1) diversification benefit, 2) consumption goals (e.g., retirement, children education, medical expenses), and 3) individual preferences (e.g., sectors, factors, themes, ESG). While diversification and consumption goals have been extensively studied within modern portfolio theory and goal-based investing, individual preferences have been less explored but now machine learning shows promise through its capacity to process vast data volumes. This presentation will introduce what each of these factors entails and discuss ongoing research aimed at crafting personalized investment strategies that take the three key factors into account.

Michael Wellman, University of Michigan, USA

Understanding the Implications of AI on Financial Markets

The rapid advancement of surprisingly capable AI is raising questions about AI's impact on virtually all aspects of our economy and society. The nexus of AI and Finance is especially salient, building on the impact AI has already had on trading and other finance-related domains. Finance may also provide some early tests in AI regulation, as existing regulatory functions may need to stretch to cover new issues posed by the latest AI advances. I discuss in particular how AI could exacerbate problems of market manipulation, and otherwise create loopholes in regulatory regimes. Finally, I suggest ways to anticipate AI impacts and evaluate capabilities of advanced interactive AI.