Jason Bohne

About Me

Hello, I’m Jason Bohne, a fourth-year Ph.D. candidate in the Department of Applied Mathematics and Statistics at Stony Brook University, advised by Professor Pawel Polak. Prior to Stony Brook, I completed a B.S. in Mathematics at the University of Illinois at Chicago from 2018 to 2021, advised by Professor Jie Yang. My expected graduation date is May 2026.

My main research interests are (i) bilevel optimization (e.g., hyperparameter optimization, meta-learning, and reinforcement learning from human feedback), and (ii) online learning (e.g., sequential optimization, adversarial learning, and regret minimizing algorithms). Recently I’ve been interested in the intersection of the aforementioned which is the novel research direction of online bilevel optimization.

I have been fortunate enough to intern with the Machine Learning Strategy Team in the CTO office at Bloomberg L.P., where I study how bilevel optimization and online learning can be applied to a variety of (financial) machine learning tasks.

In a previous chapter of my life, I was one of the first interns at Alpaca Markets where I organized two conferences on algorithmic trading and market data as well as their podcast Fintech Underground. During my undergraduate years, I co-founded and was the president of the first quantitative finance club (QTC) at UIC.

Recent News in 2024!

• [News] 12/2024 I successfully completed my preliminary exam on “Online Bilevel Optimization for Machine Learning”. Thanks to my wonderful committee!

• [Talk] 12/2024 I’ll be giving a poster on Online Nonconvex Bilevel Optimization with Bregman Divergences at the 16th International OPT Workshop on Optimization for Machine Learning at NeurIPS 2024. See you in Vancouver!

• [Talk] 10/2024 I’ll be giving a talk on online bilevel optimization at the Third Workshop on Recent Trends in Machine Learning & Risk Management at Stony Brook University.

• [Talk] 9/2024 I’ll be presenting a poster at the 10th Annual Bloomberg-Columbia Machine Learning in Finance Conference.

• [Preprint] 9/2024 Our Preprint Online Nonconvex Bilevel Optimization with Bregman Divergences is available on arXiv.

• [Teaching] 8/2024 Teaching Assistant for AMS 511: Foundation of Quantitative Finance.

• [Award] 5/2024 Awarded Excellence in Graduate Teaching Doctoral Student in Applied Mathematics and Statistics Department at Stony Brook University for teaching AMS 512: Capital Markets & Portfolio Theory.

• [Teaching] 1/2024 Instructor for AMS 512: Capital Markets & Portfolio Theory.