DJAM: Distributed Jacobi Asynchronous Method for Learning Personal Models
Open accessThe DOI matches its Crossref record; 18 of the 18 references it cites were found in a registry or an index.
Abstract
Processing data collected by a network of agents often boils down to solving an optimization problem. The distributed nature of these problems calls for methods that are, themselves, distributed. While most collaborative learning problems require agents to reach a common (or consensus) model, there are situations in which the consensus solution may not be optimal. For instance, agents may want to reach a compromise between agreeing with their neighbors and minimizing a personal loss function. We present DJAM, a Jacobi-like distributed algorithm for learning personalized models. This method is implementation-friendly: it has no hyperparameters that need tuning, it is asynchronous, and its updates only require single-neighbor interactions. We prove that DJAM converges with probability one to the solution, provided that the personal loss functions are strongly convex and have Lipschitz gradient. We then give evidence that DJAM is on par with state-of-the-art methods: our method reaches a solution with error similar to the error of a carefully tuned alternating direction method of multipliers (ADMM) in about the same number of single-neighbor interactions.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Other platforms include [125] focused on a range of medical applications and [321] for enterprise use cases.”
From Advances and Open Problems in Federated Learning (2021) · cites this paper as [125]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“- Clara Training Framework [125] includes the support of cross-silo federated learning based on a server-client approach with data privacy protection.”
From Advances and Open Problems in Federated Learning (2021) · cites this paper as [125]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Boyd, S., Parikh, N., Chu, E., Peleato, B., & Eckstein, J. (2011) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000016 · Cited by 17,075
Huber, P. J. (1981) · Wiley Series in Probability and Statistics · doi:10.1002/0471725250 · Cited by 7,066
Distributed Subgradient Methods for Multi-Agent Optimization
Nedic, A., & Ozdaglar, A. (2009) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2008.2009515 · Cited by 3,719 · cited 1×
“They use techniques ranging from distributed (sub)gradient descent [4], [5] to more elaborate techniques such as EXTRA [6], distributed ADMM [7], [8], dual averaging [9], and distributed augmented Lagrangean [10].”
MethodEXTRA: An Exact First-Order Algorithm for Decentralized Consensus Optimization
Shi, W., Ling, Q., Wu, G., & Yin, W. (2015) · SIAM Journal on Optimization · doi:10.1137/14096668x · Cited by 1,261 · cited 2×
“They use techniques ranging from distributed (sub)gradient descent [4], [5] to more elaborate techniques such as EXTRA [6], distributed ADMM [7], [8], dual averaging [9], and distributed Augmented Lagrangean (AL) [10].”
Distributed Sparse Linear Regression
Mateos, G., Bazerque, J. A., & Giannakis, G. B. (2010) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2010.2055862 · Cited by 550 · cited 1×
“Some algorithms aim at more specific optimization tasks such as distributed lasso regression [11], distributed support vector machines [12], and distributed random vector functional-link networks [13].”
MethodOn the O(1=k) convergence of asynchronous distributed alternating Direction Method of Multipliers
Wei, E., & Ozdaglar, A. (2013) · 2013 IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2013.6736937 · Cited by 342 · cited 1×
“They use techniques ranging from distributed (sub)gradient descent [4], [5] to more elaborate techniques such as EXTRA [6], distributed ADMM [7], [8], dual averaging [9], and distributed augmented Lagrangean [10].”
MethodMultitask Diffusion Adaptation Over Networks
Chen, J., Richard, C., & Sayed, A. H. (2014) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2014.2333560 · Cited by 291 · cited 1×
“Problem (1) is also referred to as multitask problem; this problem is solved in [17] for a more restricted class of personal losses than ours.”
BackgroundNetwork Lasso: Clustering and Optimization in Large Graphs
Hallac, D., Leskovec, J., & Boyd, S. (2015) · Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2783258.2783313 · Cited by 235
Consensus-based distributed linear support vector machines
Forero, P. A., Cano, A., & Giannakis, G. B. (2010) · Proceedings of the 9th ACM/IEEE International Conference on Information Processing in Sensor Networks · doi:10.1145/1791212.1791218 · Cited by 171
Parallel Selective Algorithms for Nonconvex Big Data Optimization
Facchinei, F., Scutari, G., & Sagratella, S. (2015) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2015.2399858 · Cited by 162 · cited 1×
“The methods proposed in [15] and [16] can tackle more general problems, but both require that agents communicate with all their neighbors before updating, while our method needs only communications between two agents at a time.”
Method
Cited by
Advances and Open Problems in Federated Learning
Kairouz, P., & McMahan, H. B. (2021) · Foundations and Trends in Machine Learning · doi:10.1561/2200000083 · Cited by 7,772
Adaptive Personalized Federated Learning
Deng, Y., Kamani, M. M., & Mahdavi, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.13461 · Cited by 310
“(2017); Almeida and Xavier (2018), where in addition to the optimization error, they have network constraints or peer-to-peer communication limitation (Bellet et al.”
BackgroundFully Decentralized Joint Learning of Personalized Models and Collaboration Graphs
Zantedeschi, V., Bellet, A., & Tommasi, M. (2019) · arXiv · doi:10.48550/arxiv.1901.08460 · Cited by 38
“Our work is more closely inspired by recent decentralized approaches that have shown the benefits of collaboratively learning personalized models for each user by leveraging a similarity graph given as input to the algorithm (Vanhaesebrouck et al., 2017; Li et al., 2017; Bellet et al., 2018; Almeida and Xavier, 2018).”
BackgroundMethodInfluentialPeer-to-peer deep learning with non-IID data
Šajina, R., Tanković, N., & Ipšić, I. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.119159 · Cited by 21
Distributed Second-Order Methods With Increasing Number of Working Nodes
Jerinkić, N. K., Jakovetić, D., Krejić, N., & Bajović, D. (2019) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2019.2922191 · Cited by 10
“There are several motivations for the model (1), including learning personalized models, spatial field estimation, animal flocking, etc. [34], and it has been extensively studied, e.g., [11], [12], and [34].”
BackgroundRobust P2P Personalized Learning
Boubouh, K., Boussetta, A., Benkaouz, Y., & Guerraoui, R. (2020) · IEEE Symposium on Reliable Distributed Systems (SRDS) · doi:10.1109/srds51746.2020.00037 · Cited by 12
“Several such approaches have recently been presented [3], [9], [20], [21].”
BackgroundGame-Theoretic Distributed Empirical Risk Minimization With Strategic Network Design
Liu, S., Li, T., & Zhu, Q. (2023) · IEEE Transactions on Signal and Information Processing over Networks · doi:10.1109/tsipn.2023.3306106 · Cited by 19
“Some recent endeavors include the network proximity formulation in place of consensuses [36], the multi-criterion optimization-based federated learning framework [6], the clustered federated learning framework [5], the heuristic rational optimization framework [37], the asynchronous method in [38],”
BackgroundOn the advantages of P2P ML on mobile devices
Basmadjian, R., Boubouh, K., Boussetta, A., Guerraoui, R., & Maurer, A. (2022) · ACM International Conference on Future Energy Systems (e-Energy) · doi:10.1145/3538637.3538863 · Cited by 8
“We have then: Since 𝑤 𝑘 ∈ W , which means that 𝑈 ≤ ( 1 − 𝑉 ) ∥∇ 𝐹 ( 𝑤 )∥ 2 , we also have: Let us define 𝜃 ∈ [ 0 , 2 ] such that sin 𝜃 = ∥∇ 𝐹 ( 𝑤 𝑘 )∥ .”
DiPLe: Learning Directed Collaboration Graphs for Peer-to-Peer Personalized Learning
Zheng, X., Naghizadeh, P., & Yener, A. (2022) · IEEE Information Theory Workshop (ITW) · doi:10.1109/itw54588.2022.9965838 · Cited by 5
“Peer-to-peer personalized learning has been studied much less extensively; recent work includes [4]–[6].”
BackgroundDistributed inexact Newton method with adaptive step sizes
Jakovetić, D., Krejić, N., & Malaspina, G. (2025) · Computational Optimization and Applications · doi:10.1007/s10589-025-00666-z · Cited by 5
“, [1, 5, 16, 20, 33]; see also [7] for a related formulation in federated learning settings.”
Metadata
- Authors
- Almeida, I., & Xavier, J.
- Venue
- IEEE Signal Processing Letters
- Published
- 2018-09
- Volume
- 25
- Issue
- 9
- Pages
- 1389-1392
- ISSN
- 1070-9908 (ISSN-L)
- 1558-2361
- OpenAlex
- W2794648480
- Cited by
- 10
- References
- 18
- Citations by year
- 2025: 1 · 2023: 1 · 2022: 3 · 2020: 3 · 2019: 2
- FWCI
- 0.77
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Signal Processing Letters (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Fundacio para a Ciencia e a Tecnologia (PD/BD/135012/2017, UID/EEA/50009/2013)
- Topics
- Physical Sciences › Engineering › Control and Systems Engineering › Distributed Control Multi-Agent Systems
- Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Keywords
- distributed optimization, personalized models, Jacobi method, asynchronous distributed algorithms, consensus learning, strong convexity, Lipschitz gradient, almost sure convergence, networked agents
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.