Laurent Massoulié

Timeline

1. 1991–1998

   A PhD under Pierre Brémaud

   He graduates from École Polytechnique in 1991 and earns a PhD at Université Paris-Sud in 1995 under Pierre Brémaud, with a thesis on the stability, simulation and optimization of discrete-event systems. From 1995 to 1998, he is a researcher at France Telecom R&D, where he builds mathematical models of data transport over the Internet.

2. 1999–2006

   Gossip protocols at Microsoft Research

   He joins Microsoft Research Cambridge in 1999 and works on distributed algorithms for peer-to-peer systems, especially epidemic methods for spreading information. With Ayalvadi Ganesh and Anne-Marie Kermarrec, he publishes SCAMP, a decentralized membership protocol for gossip-based group communication, in IEEE Transactions on Computers in 2003.

3. 2006–2012

   Paris Research Lab at Technicolor

   In 2006, he joins Technicolor, then known as Thomson, where he designs peer-to-peer systems for media streaming, analyzes information spread in social networks, and holds positions as director of the Paris Research Lab and distinguished scientist. With Moez Draief, he publishes the book “Epidemics and Rumours in Complex Networks” in 2010. He also completes a habilitation thesis at Université Paris 7 in 2010.

4. 2012–2014

   Directing the Microsoft Research–Inria centre

   In 2012, he joins Inria as Director of the Microsoft Research–Inria Joint Centre. When Microsoft Research and Inria renew the partnership in April 2014, Andrew Blake of Microsoft Research Cambridge credits him with extending the collaboration to machine learning and social networking.

5. November 2013

   The block model threshold

   In November 2013, his paper “Community detection thresholds and the weak Ramanujan property” solves the positive part of a conjecture by Decelle and colleagues on a sharp threshold for detecting communities in sparse graphs drawn from the stochastic block model. The proof sparks a large body of follow-up work in theoretical computer science, information theory and applied probability.

6. 2017–2018

   Optimal distributed optimization algorithms

   In 2017, with Kevin Scaman, Francis Bach, Sébastien Bubeck and Yin Tat Lee, he determines optimal convergence rates for smooth, strongly convex distributed optimization over a network. He receives the Del Duca Foundation's Grand Prix Scientifique from the French Academy of Sciences in 2017. The team's “Optimal Algorithms for Non-Smooth Distributed Optimization in Networks” wins a NeurIPS 2018 Best Paper award.

7. 2021–2023

   NeurIPS and SIGMETRICS awards

   In 2021, “Continuized Accelerations of Deterministic and Stochastic Gradient Descents, and of Gossip Algorithms,” written with Mathieu Even, Raphaël Berthier, Francis Bach, Nicolas Flammarion, Hadrien Hendrikx, Pierre Gaillard and Adrien Taylor, wins a NeurIPS Outstanding Paper award. ACM SIGMETRICS gives him its 2023 Achievement Award for contributions to the theory and practice of large-scale distributed networks and systems, including distributed optimization for federated learning.

8. As of October 2026

   Scientific delegate at Inria Paris

   He is a research director at Inria in the Argo project-team on learning, graphs and distributed optimization at the École normale supérieure, and scientific delegate of the Inria Paris centre. As a chair holder at the PRAIRIE institute, he develops distributed algorithms for learning from data spread over several machines. He is also Managing Director of the Microsoft Research–Inria Joint Centre.