Stefanie Jegelka
Papers
1
Total Citations
14
H-Index
1
About
Stefanie Jegelka is a leading researcher in machine learning, with key contributions spanning Bayesian optimization, combinatorial optimization, and the theoretical foundations of deep learning. Her work on "Optimization as Estimation with Gaussian Processes in Bandit Settings" (2015) introduced a novel perspective that reframes optimization as an estimation problem, directly leveraging Gaussian process posteriors to estimate the argmax of an unknown function. This approach has advanced Bayesian optimization, a critical tool for hyperparameter tuning and experimental design, and has been widely cited (over 14 times) for its elegant theoretical insights. Beyond this, Jegelka has made impactful strides in structured prediction, graph neural networks, and scalable algorithms for discrete optimization, often bridging theory and practice. Her research has earned her prestigious recognitions, including an NSF CAREER Award and a Sloan Research Fellowship, underscoring her influence in the field. With hundreds of citations across her body of work, Jegelka continues to shape how machines learn from limited data and complex structures, inspiring students and researchers alike to explore the intersection of optimization, probability, and learning.
Research Focus
Key Achievements
Top Papers
- 1Optimization as Estimation with Gaussian Processes in Bandit Settings14 citations · 2015