Emmanuel Abbé
Also known as Emmanuel Abbe
Emmanuel Abbé develops mathematical theory for data science and machine learning. His selected work examines neural learning, community detection and information theory. Abbé and Colin Sandon coauthored work on fundamental limits and efficient recovery algorithms for community detection in stochastic block models. It appeared at FOCS 2015. Abbé and Sandon coauthored On the Universality of Deep Learning, published at NeurIPS 2020. This is part of a broader research program on the capabilities and limits of learning algorithms. Abbé coauthored The Staircase Property with a team of researchers. The NeurIPS 2021 paper examines how hierarchical structure can guide deep learning. He also coauthored work comparing differentiable, PAC and statistical-query learning.
Timeline
-
2015
Learning communities in graphs
Abbé and Colin Sandon coauthored work on fundamental limits and efficient recovery algorithms for community detection in stochastic block models. It appeared at FOCS 2015.
-
2020
The universality of deep learning
Abbé and Sandon coauthored On the Universality of Deep Learning, published at NeurIPS 2020. This is part of a broader research program on the capabilities and limits of learning algorithms.
-
2021
Hierarchical structure in learning
Abbé coauthored The Staircase Property with a team of researchers. The NeurIPS 2021 paper examines how hierarchical structure can guide deep learning. He also coauthored work comparing differentiable, PAC and statistical-query learning.
Sources and credits
Photo: people.epfl.ch; Emmanuel Abbé profile photograph; creator not stated. License: Not stated by source. Source: people.epfl.ch. Cropped to a square and resized. Ask us to remove or replace this photo.