King Abdullah University of Science and Technology (KAUST), Saudi Arabia
Laurent Condat
Senior Research Scientist

Visual Computing Center
and AI initiative
Building 12, Office 3122



home

research

publications

talks

software

images

teaching

 

Optimization for data science: algorithms, convex relaxations, sparse and low rank models, inverse problems, image and signal processing, learning

                  
 

About me: I got my PhD in 2006 from Grenoble Inst. of Tech., Grenoble, France. After 2 years as a postdoc in Munich, Germany, I was hired as a permanent researcher by the CNRS in 2008. I spent 4 years in the GREYC, Caen, and 7 years in GIPSA-Lab, Grenoble. Since Nov. 2019, I am on leave from the CNRS and I am a researcher in the CEMSE Division of KAUST.


News

[Nov. 22] I am in Stanford's list of the world's top 2% most influential scientists.

[Oct. 22] Our paper "Provably Doubly Accelerated Federated Learning: The First Theoretically Successful Combination of Local Training and Compressed Communication" is on arXiv.

[Oct. 22] Our RandProx paper has been accepted for presentation at the NeurIPS Workshop on Optimization for Machine Learning (OPT2022).

[Sept. 22] Our paper "EF-BV: A Unified Theory of Error Feedback and Variance Reduction Mechanisms for Biased and Unbiased Compression in Distributed Optimization" has been accepted at NeurIPS 2022.

[July 22] Our paper "RandProx: Primal-Dual Optimization Algorithms with Randomized Proximal Updates" is on arXiv.

[July 22] I have been promoted to Senior Research Scientist.

[June 22] Our paper "MURANA: A Generic Framework for Stochastic Variance-Reduced Optimization" has been accepted at MSML 2022.

[May 22] My paper "Tikhonov regularization of circle-valued signals" has been accepted to IEEE Transactions on Signal Processing.

[May 22] Our paper "Dualize, split, randomize: Fast nonsmooth optimization algorithms" has been accepted to Journal of Optimization Theory and Applications.

[May 22] Our paper "EF-BV: A Unified Theory of Error Feedback and Variance Reduction Mechanisms for Biased and Unbiased Compression in Distributed Optimization", written earlier this year, is on arXiv.

[Apr. 22] Our tutorial paper "Proximal Splitting Algorithms for Convex Optimization: A Tour of Recent Advances, with New Twists" has been accepted to SIAM Review.

[Jan. 22] Our paper "Distributed Proximal Splitting Algorithms with Rates and Acceleration" has been published in Frontiers in Signal Processing.  Link

[Jan. 22] Our paper "An Optimal Algorithm for Strongly Convex Minimization under Affine Constraints" accepted at AISTATS 2022.

[Dec. 21] Our tutorial paper has been revised, again. The title has changed to "Proximal Splitting Algorithms for Convex Optimization: A Tour of Recent Advances, with New Twists".

[Nov. 21] Daniele Picone, whose work I co-supervised with Mauro Dalla Mura, defended his PhD. Congratulations Daniele!

[Apr. 21] I have been appointed as an Associate Editor for IEEE Transactions on Signal Processing.

[Feb. 21] Julien Baderot, whose work I co-supervised with Michel Desvignes and Mauro Dalla Mura, defended his PhD. Congratulations Julien!

[Nov. 20] 3 papers accepted for presentation: 2 at the NeurIPS Workshop on Optimization for Machine Learning (OPT2020) and 1 at the NeurIPS Workshop on Scalability, Privacy, and Security in Federated Learning (SpicyFL 2020).

[Oct. 20] 2 new papers on arXiv: the paper "Distributed Proximal Splitting Algorithms with Rates and Acceleration" and the paper "Optimal Gradient Compression for Distributed and Federated Learning".

[Oct. 20] Our overview paper on proximal splitting, with new relaxation results, has been revised. The title has been changed from "Splitting Algorithms: Relax them all!" to "Proximal Splitting Algorithms: A Tour of Recent Advances, with New Twists".

[July 20] My paper "Atomic norm minimization for decomposition into complex exponentials and optimal transport in Fourier domain" accepted to Journal of Approximation Theory.   Link

[June 20] Our paper "From local SGD to local fixed point methods for federated learning" accepted at ICML 2020.  Link

[May 20] Our paper on optical tomography was chosen for the cover of the journal Sensors.