paper-with-me

홈 › Papers

Federated ADMM from Bayesian Duality

2025-06-16 · Thomas Möllenhoff, Siddharth Swaroop, Finale Doshi-Velez, Mohammad Emtiyaz Khan

ADMM is a popular method for federated deep learning which originated in the 1970s and, even though many new variants of it have been proposed since then, its core algorithmic structure has remained unchanged. Here, we take a major departure from the old structure and present a fundamentally new way to derive and extend federated ADMM. We propose to use a structure called Bayesian Duality which exploits a duality of the posterior distributions obtained by solving a variational-Bayesian reformulation of the original problem. We show that this naturally recovers the original ADMM when isotropic Gaussian posteriors are used, and yields non-trivial extensions for other posterior forms. For instance, full-covariance Gaussians lead to Newton-like variants of ADMM, while diagonal covariances result in a cheap Adam-like variant. This is especially useful to handle heterogeneity in federated deep learning, giving up to 7% accuracy improvements over recent baselines. Our work opens a new Bayesian path to improve primal-dual methods.

📄 PDF Abstract BibTeX arXiv:2506.13150

Code (1)

team-approx-bayes/bayes-admm 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

ADMM The alternating direction method of multipliers (ADMM) is an algorithm that solves convex optimization problems by breaking them into smaller pieces, each of which are…

Similar Papers 제목 키워드 기반

L-FGADMM: Layer-Wise Federated Group ADMM for Communication Efficient Decentralized Deep Learning

2019-11-09 · Anis Elgabli, Jihong Park, Sabbir Ahmed, Mehdi Bennis

This article proposes a communication-efficient decentralized deep learning algorithm, coined layer-wise federated group ADMM (L-FGADMM). To minimize an empirical risk, every worker in L-FGADMM periodically communicates …

Federated Learning

Connecting Federated ADMM to Bayes

2025-01-28 · Siddharth Swaroop, Mohammad Emtiyaz Khan, Finale Doshi-Velez

We provide new connections between two distinct federated learning approaches based on (i) ADMM and (ii) Variational Bayes (VB), and propose new variants by combining their complementary strengths. Specifically, we show …

Federated Learning

Towards Federated Bayesian Network Structure Learning with Continuous Optimization

2021-10-18 · Ignavier Ng, Kun Zhang

Traditionally, Bayesian network structure learning is often carried out at a central site, in which all data is gathered. However, in practice, data may be distributed across different parties (e.g., companies, devices) …

Federated Learning

FedADMM: A Federated Primal-Dual Algorithm Allowing Partial Participation

2022-03-28 · Han Wang, Siddartha Marella, James Anderson

Federated learning is a framework for distributed optimization that places emphasis on communication efficiency. In particular, it follows a client-server broadcast model and is particularly appealing because of its abil…

Distributed OptimizationFederated Learning

Improved Communication Efficiency in Federated Natural Policy Gradient via ADMM-based Gradient Updates

2023-10-09 · NeurIPS 2023 11

Federated reinforcement learning (FedRL) enables agents to collaboratively train a global policy without sharing their individual data. However, high communication overhead remains a critical bottleneck, particularly for…

MuJoCo