Michele Pavon
Papers
2
Total Citations
107
H-Index
2
About
Michele Pavon is a leading figure in stochastic control, information theory, and optimal transport, whose work bridges deep mathematical traditions with modern data science. His most celebrated contribution centers on the Schrödinger bridge problem (SBP)—a probabilistic inference method rooted in large deviations theory, originally posed by Erwin Schrödinger in 1931–32. Pavon’s landmark paper, “Stochastic Control Liaisons: Richard Sinkhorn Meets Gaspard Monge on a Schrödinger Bridge” (2021, 102 citations), masterfully unifies three seemingly disparate fields: it reveals how the SBP can be solved via the Sinkhorn algorithm (a matrix scaling method) and serves as a regularized version of the Monge–Kantorovich optimal transport problem. This synthesis not only clarified the theoretical underpinnings of Schrödinger bridges but also provided a computationally tractable framework, sparking widespread adoption in machine learning, generative modeling, and control. Pavon’s work has profoundly shaped the modern understanding of how stochastic control, entropy regularization, and optimal transport intersect, making him a pivotal figure whose insights continue to inspire new generations of researchers in applied mathematics and engineering.
Research Focus
Key Achievements
Top Papers
- 1
- 2