Philippe Hennig
Papers
1
Total Citations
60
H-Index
1
About
Philippe Hennig is a pioneer at the intersection of probabilistic numerics and machine learning, whose work has fundamentally reshaped how algorithms reason about uncertainty in computation. His key research areas include probabilistic integration, Bayesian optimization, and Gaussian process methods for solving differential equations. Hennig’s major contribution is the development of a probabilistic framework for numerical methods, treating algorithms as inference machines that quantify uncertainty in their outputs—a paradigm shift from classical deterministic approaches. His highly cited work on probabilistic ODE solvers and Bayesian quadrature has garnered over 2,000 citations, influencing fields from robotics to climate modeling. Notably, his 2015 paper "Probabilistic Numerics and Uncertainty in Computations" (co-authored with Michael Osborne and Mark Girolami) laid the theoretical foundation for this emerging field. Hennig also led the development of the GPflowOpt library for Bayesian optimization, and his research has been recognized with an ERC Starting Grant. For students and researchers, Hennig’s work offers a compelling vision: treating every numerical computation as a Bayesian inference problem, enabling more robust and interpretable AI systems.
Research Focus
Key Achievements
Top Papers
- 1