Picture for Laurent Condat

Laurent Condat

Prune at the Clients, Not the Server: Accelerated Sparse Training in Federated Learning

May 31, 2024
Viaarxiv icon

FedComLoc: Communication-Efficient Distributed Training of Sparse and Quantized Models

Mar 14, 2024
Figure 1 for FedComLoc: Communication-Efficient Distributed Training of Sparse and Quantized Models
Figure 2 for FedComLoc: Communication-Efficient Distributed Training of Sparse and Quantized Models
Figure 3 for FedComLoc: Communication-Efficient Distributed Training of Sparse and Quantized Models
Figure 4 for FedComLoc: Communication-Efficient Distributed Training of Sparse and Quantized Models
Viaarxiv icon

LoCoDL: Communication-Efficient Distributed Learning with Local Training and Compression

Mar 07, 2024
Figure 1 for LoCoDL: Communication-Efficient Distributed Learning with Local Training and Compression
Figure 2 for LoCoDL: Communication-Efficient Distributed Learning with Local Training and Compression
Figure 3 for LoCoDL: Communication-Efficient Distributed Learning with Local Training and Compression
Figure 4 for LoCoDL: Communication-Efficient Distributed Learning with Local Training and Compression
Viaarxiv icon

RandCom: Random Communication Skipping Method for Decentralized Stochastic Optimization

Oct 12, 2023
Viaarxiv icon

Near-Linear Time Projection onto the $\ell_{1,\infty}$ Ball; Application to Sparse Autoencoders

Add code
Jul 19, 2023
Figure 1 for Near-Linear Time Projection onto the $\ell_{1,\infty}$ Ball; Application to Sparse Autoencoders
Figure 2 for Near-Linear Time Projection onto the $\ell_{1,\infty}$ Ball; Application to Sparse Autoencoders
Figure 3 for Near-Linear Time Projection onto the $\ell_{1,\infty}$ Ball; Application to Sparse Autoencoders
Figure 4 for Near-Linear Time Projection onto the $\ell_{1,\infty}$ Ball; Application to Sparse Autoencoders
Viaarxiv icon

Explicit Personalization and Local Training: Double Communication Acceleration in Federated Learning

Add code
May 22, 2023
Figure 1 for Explicit Personalization and Local Training: Double Communication Acceleration in Federated Learning
Figure 2 for Explicit Personalization and Local Training: Double Communication Acceleration in Federated Learning
Figure 3 for Explicit Personalization and Local Training: Double Communication Acceleration in Federated Learning
Figure 4 for Explicit Personalization and Local Training: Double Communication Acceleration in Federated Learning
Viaarxiv icon

TAMUNA: Accelerated Federated Learning with Local Training and Partial Participation

Feb 20, 2023
Figure 1 for TAMUNA: Accelerated Federated Learning with Local Training and Partial Participation
Figure 2 for TAMUNA: Accelerated Federated Learning with Local Training and Partial Participation
Figure 3 for TAMUNA: Accelerated Federated Learning with Local Training and Partial Participation
Viaarxiv icon

Provably Doubly Accelerated Federated Learning: The First Theoretically Successful Combination of Local Training and Compressed Communication

Oct 27, 2022
Figure 1 for Provably Doubly Accelerated Federated Learning: The First Theoretically Successful Combination of Local Training and Compressed Communication
Figure 2 for Provably Doubly Accelerated Federated Learning: The First Theoretically Successful Combination of Local Training and Compressed Communication
Figure 3 for Provably Doubly Accelerated Federated Learning: The First Theoretically Successful Combination of Local Training and Compressed Communication
Figure 4 for Provably Doubly Accelerated Federated Learning: The First Theoretically Successful Combination of Local Training and Compressed Communication
Viaarxiv icon

Joint Demosaicing and Fusion of Multiresolution Compressed Acquisitions: Image Formation and Reconstruction Methods

Add code
Sep 10, 2022
Figure 1 for Joint Demosaicing and Fusion of Multiresolution Compressed Acquisitions: Image Formation and Reconstruction Methods
Figure 2 for Joint Demosaicing and Fusion of Multiresolution Compressed Acquisitions: Image Formation and Reconstruction Methods
Figure 3 for Joint Demosaicing and Fusion of Multiresolution Compressed Acquisitions: Image Formation and Reconstruction Methods
Figure 4 for Joint Demosaicing and Fusion of Multiresolution Compressed Acquisitions: Image Formation and Reconstruction Methods
Viaarxiv icon

Tikhonov Regularization of Sphere-Valued Signals

Add code
Jul 25, 2022
Viaarxiv icon