The recent advancement of deep learning has had a profound influence on many fields, including finance. Mathematical and quantitative finance provide a plentitude of challenging prediction problems that can be used as benchmarks for deep and reinforcement learning algorithms. Specifically, financial markets represent a complex interplay of agents interacting through auction-market mechanisms at different time scales and with different objectives. Therefore, it is not surprising that it continues to receive attention from computer scientists, physicists, social scientists, and others interested in addressing a multitude of challenging prediction and decision problems. 

In this track, we will highlight recent ML advancements like transformers, physics-informed neural networks, graph neural networks, and complexity tools and their impact on decision making, data-driven analysis, and time series predictions in finance.

The purpose of this track is to enable the exchange of recent research and insights amongst researchers interested in machine learning approaches for decision making and times series analysis of financial markets. We aim to bring together world-class presenters from academia and industry working on topics such as:

• Deep learning for financial time series
• Reinforcement learning and data-driven optimal control for financial decision making
• Transformer-based and related NLP approaches for financial sentiment and event analysis
• Graph-based neural network techniques in finance

We anticipate this track will result in a most vibrant and fruitful exchange of ideas and information from researchers from different disciplines such as machine learning, complex systems, physics, mathematics and quantitative finance.