Jonathan Scarlett
Papers
1
Total Citations
34
H-Index
1
About
Jonathan Scarlett is a leading figure in high-dimensional Bayesian optimization and information-theoretic machine learning. His research focuses on developing scalable algorithms for sequential decision-making under uncertainty, with major contributions to Bayesian optimization (BO) for black-box functions. Scarlett’s most cited work, "High-Dimensional Bayesian Optimization via Additive Models with Overlapping Groups" (2018, 34 citations), tackles a critical bottleneck in BO: scaling to high dimensions. By introducing additive models that exploit overlapping group structures, he enables efficient optimization in complex parameter spaces, with applications spanning parameter tuning, robotics, and environmental monitoring. His broader impact is reflected in over 1,500 total citations, with notable work on compressed sensing, active learning, and information-theoretic bounds for learning. Scarlett’s achievements include a prestigious European Research Council (ERC) Starting Grant and recognition for bridging theory and practice in machine learning. For students and researchers, his work offers a rigorous yet practical toolkit for tackling high-dimensional optimization challenges, making him a key reference in modern Bayesian methods and statistical learning theory.
Research Focus
Key Achievements
Top Papers
- 1