In the 2024-2025 academic year, The Columbia Center of AI and Responsible Financial Innovation (CAIRFI), in collaboration with Capital One, is proud to support two faculty-led research projects and two PhD Fellows. CAIRFI is a joint effort between Columbia University and Capital One, aiming to advance AI technology in financial services and markets. Its mission is to enhance financial service operations, promote responsible AI, assess risk, improve forecasting and testing, and more.
Columbia EE is excited to announce that PhD student Leonardo Toso, advised by Professor James Anderson, was selected to be the fellowship awardee for his project “Bayesian Priors for Efficient Multi-task Representation Learning.”
About the Project: Bayesian Priors for Efficient Multi-task Representation Learning
Leonardo Toso's research focuses on learning latent representations from multi-task, non-i.i.d., and non-isotropic datasets while leveraging prior information on local and global latent variables to enhance the recovery process. A key idea in recent machine learning advancements is extracting shared features from diverse task data. Utilizing all available data to unveil a latent representation across multiple tasks reduces computational complexity and improves statistical generalization by minimizing the number of parameters needing fine-tuning for a specific task. This is particularly relevant for making accurate financial portfolio recommendations based on a client's personal investment preferences. Multiple clients in a database may share common interests, making it crucial to learn such representations to perform accurate and efficient predictions on clients' predilections to meet their long-term financial objectives. Additionally, prior information on the representation (e.g., sparsity, low-rankness, structural information, engineer's knowledge) can be crucial for a more efficient multi-task representation learning framework.
About Leonardo Toso
Leonardo Toso is a second-year Ph.D. student in the Department of Electrical Engineering at Columbia University, advised by Professor James Anderson. His research interests lie at the intersection of control theory, machine learning, and optimization. Before joining Columbia, he was an undergraduate research assistant in the Department of Engineering Science at the University of Oxford. Leonardo holds an M.S. in Control, Signal, and Image Processing from the University of Paris-Saclay, an M.Eng. in Electrical Engineering from CentraleSupélec, and a B.Eng. in Electrical Engineering from the University of Campinas (Unicamp).
