Abstract
Typical federated learning (FL) problem formulation requires learning a single model suitable for all parties while prohibiting parties from sharing their data with the aggregator. However, it may not be possible to learn a common single model that is suitable for all parties. For example, consider a sentence completion problem: “I live in the state of …” The answer clearly depends on the party, and no single model is appropriate here. To handle such situations, various personalization strategies have been proposed in the recent literature. In particular, the problem appears to have a close connection to meta-learning. We review recent FL personalization techniques categorizing them into eight groups and summarize three strategies and corresponding datasets for benchmarking personalization in federated learning. We provide an overview of the statistical challenges of personalization in federated learning. At a high level, personalization leads to an increase in the model complexity, which in turn increases the hardness of the federated learning task. We study when too much personalization can prevent standard approaches to personalized federated learning from learning the common parts of the parties and present alternative approaches that overcome such issues.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Amir S, Wallace BC, Lyu H, Silva PCMJ (2016) Modelling context with user embeddings for sarcasm detection in social media. arXiv preprint arXiv:160700976
Ammad-Ud-Din M, Ivannikova E, Khan SA, Oyomno W, Fu Q, Tan KE, Flanagan A (2019) Federated collaborative filtering for privacy-preserving personalized recommendation system. arXiv preprint arXiv:190109888
Arivazhagan MG, Aggarwal V, Singh AK, Choudhary S (2019) Federated learning with personalization layers. arXiv preprint arXiv:191200818
Baxter J (2000) A model of inductive bias learning. J Artif Intell Res 12:149–198
Bui D, Malik K, Goetz J, Liu H, Moon S, Kumar A, Shin KG (2019) Federated user representation learning. arXiv preprint arXiv:190912535
Caldas S, Duddu SMK, Wu P, Li T, Konečnỳ J, McMahan HB, Smith V, Talwalkar A (2018) Leaf: a benchmark for federated settings. arXiv preprint arXiv:181201097
Caruana R (1997) Multitask learning. Mach Learn 28(1):41–75
Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) Smote: synthetic minority over-sampling technique. J Artif Intell Res 16:321–357
Chen F, Luo M, Dong Z, Li Z, He X (2018) Federated meta-learning with fast convergence and efficient communication. arXiv preprint arXiv:180207876
Chen M, Suresh AT, Mathews R, Wong A, Allauzen C, Beaufays F, Riley M (2019) Federated learning of n-gram language models. In: Proceedings of the 23rd conference on computational natural language learning (CoNLL), pp 121–130
Cheng Y, Wang D, Zhou P, Zhang T (2017) A survey of model compression and acceleration for deep neural networks. arXiv preprint arXiv:171009282
Chernozhukov V, Chetverikov D, Demirer M, Duflo E, Hansen C, Newey W, Robins J (2017) Double/debiased machine learning for treatment and causal parameters. arXiv:160800060 [econ, stat] 1608.00060
Cohen G, Afshar S, Tapson J, Van Schaik A (2017) EMNIST: extending MNIST to handwritten letters. In: 2017 international joint conference on neural networks (IJCNN). IEEE, pp 2921–2926
Corinzia L, Beuret, A, Buhmann JM (2019) Variational federated multi-task learning. arXiv preprint arXiv:190606268
Deng Y, Kamani MM, Mahdavi M (2020) Adaptive personalized federated learning. arXiv preprint arXiv:200313461
Dinh CT, Tran NH, Nguyen TD (2020) Personalized federated learning with Moreau envelopes. arXiv preprint arXiv:200608848
Fallah A, Mokhtari A, Ozdaglar A (2020) Personalized federated learning: a meta-learning approach. arXiv preprint arXiv:200207948
Finn C, Abbeel P, Levine S (2017) Model-agnostic meta-learning for fast adaptation of deep networks. In: International conference on machine learning. PMLR, pp 1126–1135
Ghosh A, Hong J, Yin D, Ramchandran K (2019) Robust federated learning in a heterogeneous environment. arXiv preprint arXiv:190606629
Ghosh A, Chung J, Yin D, Ramchandran K (2020) An efficient framework for clustered federated learning. arXiv preprint arXiv:200604088
Gou J, Yu B, Maybank SJ, Tao D (2020) Knowledge distillation: a survey. arXiv preprint arXiv:200605525
Grbovic M, Cheng H (2018) Real-time personalization using embeddings for search ranking at Airbnb. In: Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining, pp 311–320
Hanzely F, Richtárik P (2020) Federated learning of a mixture of global and local models. arXiv preprint arXiv:200205516
Hinton G, Vinyals O, Dean J (2015) Distilling the knowledge in a neural network. arXiv preprint arXiv:150302531
Hospedales T, Antoniou A, Micaelli P, Storkey A (2020) Meta-learning in neural networks: a survey. arXiv preprint arXiv:200405439
Huang Y, Chu L, Zhou Z, Wang L, Liu J, Pei J, Zhang Y (2021) Personalized cross-silo federated learning on non-IID data. In: Proceedings of the AAAI conference on artificial intelligence, vol 35, pp 7865–7873
Jaech A, Ostendorf M (2018) Personalized language model for query auto-completion. arXiv preprint arXiv:180409661
Jeong E, Oh S, Kim H, Park J, Bennis M, Kim SL (2018) Communication-efficient on-device machine learning: federated distillation and augmentation under non-IID private data. arXiv preprint arXiv:181111479
Jiang Y, Konečnỳ J, Rush K, Kannan S (2019) Improving federated learning personalization via model agnostic meta learning. arXiv preprint arXiv:190912488
Johnson AE, Pollard TJ, Shen L, Li-Wei HL, Feng M, Ghassemi M, Moody B, Szolovits P, Celi LA, Mark RG (2016) MIMIC-III, a freely accessible critical care database. Sci Data 3(1):1–9
Khodak M, Balcan MF, Talwalkar A (2019) Adaptive gradient-based meta-learning methods. arXiv preprint arXiv:190602717
Krizhevsky A, Hinton G et al (2009) Learning multiple layers of features from tiny images
Lancaster T (2000) The incidental parameter problem since 1948. J Econ 95(2):391–413. https://doi.org/10.1016/S0304-4076(99)00044-5
Li D, Wang J (2019) FedMD: Heterogenous federated learning via model distillation. arXiv preprint arXiv:191003581
Liang PP, Liu T, Ziyin L, Allen NB, Auerbach RP, Brent D, Salakhutdinov R, Morency LP (2020) Think locally, act globally: federated learning with local and global representations. arXiv preprint arXiv:200101523
Mansour Y, Mohri M, Ro J, Suresh AT (2020) Three approaches for personalization with applications to federated learning. arXiv preprint arXiv:200210619
Mariani G, Scheidegger F, Istrate R, Bekas C, Malossi C (2018) BAGAN: data augmentation with balancing GAN. arXiv preprint arXiv:180309655
McGraw I, Prabhavalkar R, Alvarez R, Arenas MG, Rao K, Rybach D, Alsharif O, Sak H, Gruenstein A, Beaufays F et al (2016) Personalized speech recognition on mobile devices. In: 2016 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, pp 5955–5959
McMahan B, Moore E, Ramage D, Hampson S, y Arcas BA (2017) Communication-efficient learning of deep networks from decentralized data. In: Artificial intelligence and statistics. PMLR, pp 1273–1282
Mikołajczyk A, Grochowski M (2018) Data augmentation for improving deep learning in image classification problem. In: 2018 international interdisciplinary PhD workshop (IIPhDW). IEEE, pp 117–122
Minka TP (2013) Expectation propagation for approximate Bayesian inference. arXiv preprint arXiv:13012294
Neyman J (1979) C(α) tests and their use. Sankhyā: Indian J Stat Ser A (1961–2002) 41(1/2):1–21
Nichol A, Achiam J, Schulman J (2018) On first-order meta-learning algorithms. arXiv:180302999 [cs] 1803.02999
Peterson D, Kanani P, Marathe VJ (2019) Private federated learning with domain adaptation. arXiv preprint arXiv:191206733
Raghu A, Raghu M, Bengio S, Vinyals O (2019) Rapid learning or feature reuse? Towards understanding the effectiveness of MAML. In: International conference on learning representations
Ruder S (2017) An overview of multi-task learning in deep neural networks. arXiv preprint arXiv:170605098
Sattler F, Müller KR, Samek W (2020) Clustered federated learning: model-agnostic distributed multitask optimization under privacy constraints. IEEE Trans Neural Netw Learn Syst 1–13. https://doi.org/10.1109/TNNLS.2020.3015958
Schafer JB, Frankowski D, Herlocker J, Sen S (2007) Collaborative filtering recommender systems. In: Brusilovsky P, Kobsa A, Nejdl W (eds) The adaptive web: methods and strategies of web personalization. Springer, pp 291–324. https://doi.org/10.1007/978-3-540-72079-9_9
Shamir O, Srebro N, Zhang T (2014) Communication-efficient distributed optimization using an approximate Newton-type method. In: International conference on machine learning. PMLR, pp 1000–1008
Shen T, Zhang J, Jia X, Zhang F, Huang G, Zhou P, Wu F, Wu C (2020) Federated mutual learning. arXiv preprint arXiv:200616765
Shin M, Hwang C, Kim J, Park J, Bennis M, Kim SL (2020) XOR mixup: privacy-preserving data augmentation for one-shot federated learning. arXiv preprint arXiv:200605148
Shorten C, Khoshgoftaar TM (2019) A survey on image data augmentation for deep learning. J Big Data 6(1):1–48
Smith V, Chiang CK, Sanjabi M, Talwalkar A (2017) Federated multi-task learning. arXiv preprint arXiv:170510467
Thrun S (1996) Is learning the n-th thing any easier than learning the first? In: Advances in neural information processing systems. Morgan Kaufmann Publishers, San Mateo, pp 640–646
Thrun S (1998) Lifelong learning algorithms. In: Learning to learn. Springer, Boston pp 181–209
Vanschoren J (2018) Meta-learning: a survey. arXiv preprint arXiv:181003548
Vilalta R, Drissi Y (2002) A perspective view and survey of meta-learning. Artif Intell Rev 18(2):77–95
Vu T, Nguyen DQ, Johnson M, Song D, Willis A (2017) Search personalization with embeddings. In: European conference on information retrieval. Springer, pp 598–604
Weng L (2018) Meta-learning: learning to learn fast. lilianwenggithubio/lil-log. http://lilianweng.github.io/lil-log/2018/11/29/meta-learning.html
Weyand T, Araujo A, Cao B, Sim J (2020) Google Landmarks Dataset v2-a large-scale benchmark for instance-level recognition and retrieval. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 2575–2584
Yu P, Kundu A, Wynter L, Lim SH (2021) Fed+: a unified approach to robust personalized federated learning. 2009.06303
Yu T, Bagdasaryan E, Shmatikov V (2020) Salvaging federated learning by local adaptation. arXiv preprint arXiv:200204758
Yuksel SE, Wilson JN, Gader PD (2012) Twenty years of mixture of experts. IEEE Trans Neural Netw Learn Syst 23(8):1177–1193
Yurochkin M, Agarwal M, Ghosh S, Greenewald K, Hoang N, Khazaeni Y (2019) Bayesian nonparametric federated learning of neural networks. In: International conference on machine learning. PMLR, pp 7252–7261
Zhang Y, Yang Q (2017) A survey on multi-task learning. arXiv preprint arXiv:170708114
Zhao Y, Li M, Lai L, Suda N, Civin D, Chandra V (2018) Federated learning with non-IID data. arXiv preprint arXiv:180600582
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Agarwal, M., Yurochkin, M., Sun, Y. (2022). Personalization in Federated Learning. In: Ludwig, H., Baracaldo, N. (eds) Federated Learning. Springer, Cham. https://doi.org/10.1007/978-3-030-96896-0_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-96896-0_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-96895-3
Online ISBN: 978-3-030-96896-0
eBook Packages: Computer ScienceComputer Science (R0)