L. Bottou

About

Léon Bottou is a pioneering figure in machine learning, best known for his foundational contributions to stochastic gradient descent (SGD) and large-scale learning. His work on SGD, particularly the development of the "Stochastic Gradient Langevin Dynamics" and the "Bottou-Curtis-Nocedal" method, has become a cornerstone of modern deep learning, enabling efficient training of neural networks on massive datasets. With over 100,000 citations across his career, Bottou's research spans online learning, optimization theory, and natural language processing. He is also recognized for his influential work on the "Curse of Dimensionality" in non-parametric regression, as seen in his paper on Riemannian manifolds (2009, 4 citations), which addresses challenges in signal processing and computer vision. A former researcher at Microsoft Research and now at Facebook AI Research, Bottou has received numerous accolades, including the 2021 IEEE Neural Networks Pioneer Award. His practical algorithms and theoretical insights continue to shape the landscape of artificial intelligence, making him a key reference for students and researchers alike.

Research Focus

Key Achievements

Top Papers

1. 1
   Non-parametric Regression between Riemannian Manifolds

   4 citations · 2009

Key Collaborators

Contact & Links