Salvaging Federated Learning by Local Adaptation

• URL: http://arxiv.org/abs/2002.04758v3
• Date: Thu, 3 Mar 2022 23:28:58 GMT
• Title: Salvaging Federated Learning by Local Adaptation
• Authors: Tao Yu, Eugene Bagdasaryan, Vitaly Shmatikov
• Abstract summary: Federated learning (FL) is a heavily promoted approach for training ML models on sensitive data. We look at FL from the emphlocal viewpoint of an individual participant and ask: do participants have an incentive to participate in FL? We show that on standard tasks such as next-word prediction, many participants gain no benefit from FL because the federated model is less accurate on their data than the models they can train locally on their own. We evaluate three techniques for local adaptation of federated models: fine-tuning, multi-task learning, and knowledge distillation.
• Score: 26.915147034955925
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Federated learning (FL) is a heavily promoted approach for training ML models on sensitive data, e.g., text typed by users on their smartphones. FL is expressly designed for training on data that are unbalanced and non-iid across the participants. To ensure privacy and integrity of the fedeated model, latest FL approaches use differential privacy or robust aggregation. We look at FL from the \emph{local} viewpoint of an individual participant and ask: (1) do participants have an incentive to participate in FL? (2) how can participants \emph{individually} improve the quality of their local models, without re-designing the FL framework and/or involving other participants? First, we show that on standard tasks such as next-word prediction, many participants gain no benefit from FL because the federated model is less accurate on their data than the models they can train locally on their own. Second, we show that differential privacy and robust aggregation make this problem worse by further destroying the accuracy of the federated model for many participants. Then, we evaluate three techniques for local adaptation of federated models: fine-tuning, multi-task learning, and knowledge distillation. We analyze where each is applicable and demonstrate that all participants benefit from local adaptation. Participants whose local models are poor obtain big accuracy improvements over conventional FL. Participants whose local models are better than the federated model\textemdash and who have no incentive to participate in FL today\textemdash improve less, but sufficiently to make the adapted federated model better than their local models.

Related papers

• GAI-Enabled Explainable Personalized Federated Semi-Supervised Learning [29.931169585178818]
  Federated learning (FL) is a commonly distributed algorithm for mobile users (MUs) training artificial intelligence (AI) models. We propose an explainable personalized FL framework, called XPFL. Particularly, in local training, we utilize a generative AI (GAI) model to learn from large unlabeled data. In global aggregation, we obtain the new local local model by fusing the local and global FL models in specific proportions. Finally, simulation results validate the effectiveness of the proposed XPFL framework.
  arXiv  Detail & Related papers  (2024-10-11T08:58:05Z)
• A Survey on Efficient Federated Learning Methods for Foundation Model Training [62.473245910234304]
  Federated Learning (FL) has become an established technique to facilitate privacy-preserving collaborative training across a multitude of clients. In the wake of Foundation Models (FM), the reality is different for many deep learning applications. We discuss the benefits and drawbacks of parameter-efficient fine-tuning (PEFT) for FL applications.
  arXiv  Detail & Related papers  (2024-01-09T10:22:23Z)
• Tunable Soft Prompts are Messengers in Federated Learning [55.924749085481544]
  Federated learning (FL) enables multiple participants to collaboratively train machine learning models using decentralized data sources. The lack of model privacy protection in FL becomes an unneglectable challenge. We propose a novel FL training approach that accomplishes information exchange among participants via tunable soft prompts.
  arXiv  Detail & Related papers  (2023-11-12T11:01:10Z)
• VFedMH: Vertical Federated Learning for Training Multiple Heterogeneous Models [53.30484242706966]
  This paper proposes a novel approach called Vertical federated learning for training multiple Heterogeneous models (VFedMH) To protect the participants' local embedding values, we propose an embedding protection method based on lightweight blinding factors. Experiments are conducted to demonstrate that VFedMH can simultaneously train multiple heterogeneous models with heterogeneous optimization and outperform some recent methods in model performance.
  arXiv  Detail & Related papers  (2023-10-20T09:22:51Z)
• Towards More Suitable Personalization in Federated Learning via Decentralized Partial Model Training [67.67045085186797]
  Almost all existing systems have to face large communication burdens if the central FL server fails. It personalizes the "right" in the deep models by alternately updating the shared and personal parameters. To further promote the shared parameters aggregation process, we propose DFed integrating the local Sharpness Miniization.
  arXiv  Detail & Related papers  (2023-05-24T13:52:18Z)
• Can Fair Federated Learning reduce the need for Personalisation? [9.595853312558276]
  Federated Learning (FL) enables training ML models on edge clients without sharing data. This paper evaluates two Fair FL (FFL) algorithms as starting points for personalisation. We propose Personalisation-aware Federated Learning (PaFL) as a paradigm that pre-emptively uses personalisation losses during training.
  arXiv  Detail & Related papers  (2023-05-04T11:03:33Z)
• PromptFL: Let Federated Participants Cooperatively Learn Prompts Instead of Models -- Federated Learning in Age of Foundation Model [23.916918530195826]
  We propose a brand-new FL framework, PromptFL, that replaces the federated model training with the federated prompt training. PromptFL ships an off-the-shelf FM, i.e., CLIP, to distributed clients who would cooperatively train shared soft prompts. We empirically analyze the PromptFL via extensive experiments, and show its superiority in terms of system feasibility, user privacy, and performance.
  arXiv  Detail & Related papers  (2022-08-24T15:50:58Z)
• FedMR: Fedreated Learning via Model Recombination [7.404225808071622]
  Federated Learning (FL) enables global model training across clients without compromising their confidential local data. Existing FL methods rely on Federated Averaging (FedAvg)-based aggregation. This paper proposes a novel and effective FL paradigm named FedMR (Federating Model Recombination)
  arXiv  Detail & Related papers  (2022-08-16T11:30:19Z)
• FL Games: A federated learning framework for distribution shifts [71.98708418753786]
  Federated learning aims to train predictive models for data that is distributed across clients, under the orchestration of a server. We propose FL Games, a game-theoretic framework for federated learning for learning causal features that are invariant across clients.
  arXiv  Detail & Related papers  (2022-05-23T07:51:45Z)
• Personalized Federated Learning with Clustered Generalization [16.178571176116073]
  We study the recent emerging personalized learning (PFL) that aims at dealing with the challenging problem of Non-I.I.D. data in the learning setting. Key difference between PFL and conventional FL methods in the training target. We propose a novel concept called clustered generalization to handle the challenge of statistical heterogeneity in FL.
  arXiv  Detail & Related papers  (2021-06-24T14:17:00Z)
• Federated Robustness Propagation: Sharing Adversarial Robustness in Federated Learning [98.05061014090913]
  Federated learning (FL) emerges as a popular distributed learning schema that learns from a set of participating users without requiring raw data to be shared. adversarial training (AT) provides a sound solution for centralized learning, extending its usage for FL users has imposed significant challenges. We show that existing FL techniques cannot effectively propagate adversarial robustness among non-iid users. We propose a simple yet effective propagation approach that transfers robustness through carefully designed batch-normalization statistics.
  arXiv  Detail & Related papers  (2021-06-18T15:52:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.