FDML: A Collaborative Machine Learning Framework for Distributed Features
The DOI matches its Crossref record; 24 of the 26 references it cites were found in a registry or an index.
Abstract
Most current distributed machine learning systems try to scale up model training by using a data-parallel architecture that divides the computation for different samples among workers. We study distributed machine learning from a different motivation, where the information about the same samples, e.g., users and objects, are owned by several parities that wish to collaborate but do not want to share raw data with each other. We propose an asynchronous stochastic gradient descent (SGD) algorithm for such a feature distributed machine learning (FDML) problem, to jointly learn from distributed features, with theoretical convergence guarantees under bounded asynchrony. Our algorithm does not require sharing the original features or even local model parameters between parties, thus preserving the data locality. The system can also easily incorporate differential privacy mechanisms to preserve a higher level of privacy. We implement the FDML system in a parameter server architecture and compare our system with fully centralized learning (which violates data locality) and learning based on only local features, through extensive experiments performed on both a public data set a9a, and a large dataset of 5,000,000 records and 8700 decentralized features from three collaborating apps at Tencent including Tencent MyApp, Tecent QQ Browser and Tencent Mobile Safeguard. Experimental results have demonstrated that the proposed FDML system can be used to significantly enhance app recommendat…
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“Therefore, the data features and the associated model parametric blocks are split among edge devices, for which the asynchronous SGD method can be applied for vertical FL [157].”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [157]
Citations
Cites
Communication-Efficient Learning of Deep Networks from Decentralized Data
McMahan, H. B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. Y. (2016) · arXiv · doi:10.48550/arxiv.1602.05629 · Cited by 14,215 · cited 1×
“As a result, a number of distributed machine learning techniques have been proposed to collaboratively train a model by letting each party perform local model updates and exchange locally computed gradients [27] or model parameters [21] with the central server to iteratively improve model accuracy.”
BackgroundThe Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2014) · Foundations and Trends® in Theoretical Computer Science · doi:10.1561/0400000042 · Cited by 5,557 · cited 2×
“Moreover, we further enhance the data privacy by adopting a differential-privacy-based method [9, 10, 27].”
MethodLarge Scale Distributed Deep Networks
Dean, J. B., Corrado, G. S., Monga, R., Chen, K., Devin, M., Mao, M., Ranzato, M., Senior, A., Tucker, P. A., Yang, K., et al. (2012) · neural information processing systems · Cited by 4,489
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2013) · doi:10.1561/9781601988195 · Cited by 3,856
Differential Privacy: A Survey of Results
Dwork, C. (2008) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-79228-4_1 · Cited by 3,542 · cited 2×
“Moreover, we further enhance the data privacy by adopting a differential-privacy-based method [9, 10, 27].”
MethodPrivacy-preserving deep learning
Shokri, R., & Shmatikov, V. (2015) · Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2810103.2813687 · Cited by 2,356 · cited 6×
“To further protect the feature data at each party from malicious servers and parties, we apply differential privacy based methods by perturbing the local predictions to be uploaded [9, 10, 27].”
BackgroundMethodInfluentialSecureML: A System for Scalable Privacy-Preserving Machine Learning
Mohassel, P., & Zhang, Y. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.12 · Cited by 1,936 · cited 1×
“, via homomorphic encryption [11, 28] or secret sharing [3, 22, 29].”
BackgroundDeep Models Under the GAN: Information Leakage from Collaborative Deep Learning
Hitaj, B., Ateniese, G., & Perez-Cruz, F. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3134012 · Cited by 1,377 · cited 2×
“Although raw data are not transferred by the distributed SGD in [27], a recent study [12] points out that an algorithm that passes model parameters may be vulnerable to model inversion attacks based on generative adversarial networks (GANs).”
Background(2016) · International Conference on Machine Learning · Cited by 1,270
Hogwild: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent
Niu, F., Recht, B., Re, C., & Wright, S. J. (2011) · arXiv · doi:10.48550/arxiv.1106.5730 · Cited by 1,211 · cited 1×
“The convergence results have been developed for both gradient descent and SGD [13, 20, 25] as well as proximal gradient methods [18] under different assumptions of the loss functions.”
MethodParallelized Stochastic Gradient Descent
Zinkevich, M., Weimer, M., Li, L., & Smola, A. J. (2010) · Cited by 1,124
Scaling distributed machine learning with the parameter server
Li, M., Andersen, D. G., Park, J. W., Smola, A. J., Ahmed, A. A., Josifovski, V., Long, J. E., Shekita, E. J., & Su, B.-Y. (2014) · Operating Systems Design and Implementation · doi:10.5555/2685048.2685095 · Cited by 1,021
Communication Efficient Distributed Machine Learning with the Parameter Server
Li, M., Andersen, D. G., Smola, A. J., & Yu, K. (2014) · neural information processing systems · Cited by 715
Project Adam: building an efficient and scalable deep learning training system
Chilimbi, T. M., Suzue, Y., Apacible, J., & Kalyanaraman, K. (2014) · Operating Systems Design and Implementation · doi:10.5555/2685048.2685094 · Cited by 550
Optimal Distributed Online Prediction using Mini-Batches
Dekel, O., Gilad-Bachrach, R., Shamir, O., & Xiao, L. (2010) · arXiv (Cornell University) · doi:10.48550/arxiv.1012.1367 · Cited by 512 · cited 1×
“Bulk synchronous parallel algorithms (BSP) [7, 32] are among the first distributed machine learning algorithms.”
BackgroundMore Effective Distributed ML via a Stale Synchronous Parallel Parameter Server
Ho, Q., Cipar, J., Cui, H., Kim, J. K., Lee, S., Gibbons, P. B., Gibson, G. A., Ganger, G. R., & Xing, E. P. (2013) · PubMed · doi:10.1184/r1/6475898 · Cited by 506
Multiparty Differential Privacy via Aggregation of Locally Trained Classifiers
Pathak, M. A., Rane, S. D., & Raj, B. (2010) · neural information processing systems · Cited by 204
Langford, J. C., Smola, A. J., & Zinkevich, M. (2009) · arXiv (Cornell University) · doi:10.48550/arxiv.0911.0491 · Cited by 178
Strategies and Principles of Distributed Machine Learning on Big Data
Xing, E. P., Ho, Q., Xie, P., & Wei, D. (2016) · Engineering · doi:10.1016/j.eng.2016.02.008 · Cited by 162
(2012) · NOT FOUND REPOSITORY (Indian Institute of Science Bangalore) · Cited by 93
On Model Parallelization and Scheduling Strategies for Distributed Machine Learning
Lee, S., Kim, J. K., Zheng, X., Ho, Q., Gibson, G. A., & P Xing, E. (2018) · Figshare · doi:10.1184/r1/6476048 · Cited by 87
Privacy-preservation for gradient descent methods
Wan, L., Ng, W. K., Han, S., & Lee, V. C. S. (2007) · Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/1281192.1281275 · Cited by 85 · cited 1×
“, via homomorphic encryption [11, 28] or secret sharing [3, 22, 29].”
BackgroundDiFacto: Distributed Factorization Machines
Li, M., Liu, Z., Smola, A. J., & Wang, Y.-X. (2016) · Proceedings of the Ninth ACM International Conference on Web Search and Data Mining · doi:10.1145/2835776.2835781 · Cited by 47 · cited 3×
“A similar bounded delay condition is enforced in most Parameter-Server-like systems [1, 5, 14, 17, 19] to ensure convergence and avoid chaotic behavior of a completely asynchronous system.”
BackgroundOn Convergence of Model Parallel Proximal Gradient Algorithm for Stale Synchronous Parallel System
Zhou, Y., Yu, Y., Dai, W., Liang, Y., & Xing, E. P. (2016) · International Conference on Artificial Intelligence and Statistics · Cited by 32
Research works printed but not linked: 14
- Abadi Mart'in, 2016, Proc. USENIX Symposium on Operating System Design and Implementation (OSDI) .
- Ben-Nun Tal, 2018, Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency Analysis. arXiv preprint arXiv:1802.09941 · arXiv
- Bradley Joseph K, 2011, Parallel coordinate descent for l1-regularized loss minimization. arXiv preprint arXiv:1105.5379 · arXiv
- Dean Jeffrey, 2012, et almbox
- 10.5555/2503308.2188391 · doi:10.5555/2503308.2188391
- 10.5555/1791834.1791836 · doi:10.5555/1791834.1791836
- Gilad-Bachrach Ran, 2016, International Conference on Machine Learning . 201--210
- Ho Qirong, 2013, Phillip B Gibbons, Garth A Gibson, Greg Ganger, and Eric P Xing.
- Hsieh Kevin, 2017, Gaia: Geo-Distributed Machine Learning Approaching LAN Speeds.. In NSDI. 629--647.
- Lee Seunghak, 2014, Xun Zheng, Qirong Ho, Garth A Gibson, and Eric P Xing.
- Xiangru Lian Yijun Huang Yuncheng Li and Ji Liu. 2015. Asynchronous parallel stochastic gradient for nonconvex optimization. In Advances in Neural Information Processing Systems. 2737--2745. Xiangru Lian Yijun Huang Yuncheng Li and Ji Liu. 2015. Asynchronous parallel stochastic gradient for nonconvex optimization. In Advances in Neural Information Processing Systems. 2737--2745.
- McMahan H Brendan, 2016, et almbox
- Chad Scherrer Ambuj Tewari Mahantesh Halappanavar and David Haglin. 2012. Feature clustering for accelerating parallel coordinate descent. In Advances in Neural Information Processing Systems. 28--36. Chad Scherrer Ambuj Tewari Mahantesh Halappanavar and David Haglin. 2012. Feature clustering for accelerating parallel coordinate descent. In Advances in Neural Information Processing Systems. 28--36.
- Takabi Hassan, 2016, 29th Annual Conference on Neural Information Processing Systems (NIPS) .
Cited by
A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection
Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., Li, Y., Liu, X., & He, B. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3124599 · Cited by 1,407
“[71] propose FDML for vertical FL assuming all parties have the labels.”
BackgroundEdge artificial intelligence for 6g: vision, enabling technologies, and applications
Letaief, K. B., Shi, Y., Lu, J., & Lu, J. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3126076 · Cited by 829
“Therefore, the data features and the associated model parametric blocks are split among edge devices, for which the asynchronous SGD method can be applied for vertical FL [157].”
MethodVertical Federated Learning: Concepts, Advances, and Challenges
Liu, Y., Kang, Y., Zou, T., Pu, Y., He, Y., Ye, X., Ouyang, Y., Zhang, Y.-Q., & Yang, Q. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3352628 · Cited by 378
“[27] proposed FDML, allowing each party to update its local model asynchronously but based on the same sequence of randomly sampled training data.”
BackgroundMethodInfluentialA Survey of Incentive Mechanism Design for Federated Learning
Zhan, Y., Zhang, J., Hong, Z., Wu, L., Li, P., & Guo, S. (2021) · IEEE Transactions on Emerging Topics in Computing · doi:10.1109/tetc.2021.3063517 · Cited by 284
FedBCD: A Communication-Efficient Collaborative Learning Framework for Distributed Features
Liu, Y., Zhang, X., Kang, Y., Li, L., Chen, T., Hong, M., & Yang, Q. (2022) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2022.3198176 · Cited by 257
“Feature-partitionedcollaborativelearningproblemshave been studied in the setting of both DL [4], [5], [6] and FL [7], [8], [9], [10].”
Background
Metadata
- Authors
- Hu, Y., Niu, D., Yang, J., & Zhou, S.
- Affiliations
- University of Alberta (CA)
- Tencent (China) (CN)
- Venue
- Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
- Publisher
- ACM
- Published
- 2019-07-25
- Pages
- 2232-2240
- OpenAlex
- W2951059495
- Cited by
- 163
- References
- 26
- Citations by year
- 2026: 7 · 2025: 21 · 2024: 29 · 2023: 30 · 2022: 33 · 2021: 28 · 2020: 12 · 2019: 3
- FWCI
- 7.27
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Computer Networks and Communications › Cryptography and Data Security
- Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
- Keywords
- distributed learning, privacy-preserving machine learning, data ownership
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.