paper-with-me

홈 › Papers

Initialization Matters: Unraveling the Impact of Pre-Training on Federated Learning

2025-02-11 · Divyansh Jhunjhunwala, Pranay Sharma, Zheng Xu, Gauri Joshi

Initializing with pre-trained models when learning on downstream tasks is becoming standard practice in machine learning. Several recent works explore the benefits of pre-trained initialization in a federated learning (FL) setting, where the downstream training is performed at the edge clients with heterogeneous data distribution. These works show that starting from a pre-trained model can substantially reduce the adverse impact of data heterogeneity on the test performance of a model trained in a federated setting, with no changes to the standard FedAvg training algorithm. In this work, we provide a deeper theoretical understanding of this phenomenon. To do so, we study the class of two-layer convolutional neural networks (CNNs) and provide bounds on the training error convergence and test error of such a network trained with FedAvg. We introduce the notion of aligned and misaligned filters at initialization and show that the data heterogeneity only affects learning on misaligned filters. Starting with a pre-trained model typically results in fewer misaligned filters at initialization, thus producing a lower test error even when the model is trained in a federated setting with data heterogeneity. Experiments in synthetic settings and practical FL training on CNNs verify our theoretical findings.

📄 PDF Abstract BibTeX arXiv:2502.08024

Code (0)

등록된 구현이 없습니다.

Tasks

Federated Learning

Similar Papers 제목 키워드 기반

Where to Begin? On the Impact of Pre-Training and Initialization in Federated Learning

2022-10-14 · John Nguyen, Jianyu Wang, Kshitiz Malik, Maziar Sanjabi 외

An oft-cited challenge of federated learning is the presence of heterogeneity. \emph{Data heterogeneity} refers to the fact that data from different clients may follow very different distributions. \emph{System heterogen…

Federated Learning

Where to Begin? On the Impact of Pre-Training and Initialization in Federated Learning

2022-06-30 · John Nguyen, Jianyu Wang, Kshitiz Malik, Maziar Sanjabi 외

An oft-cited challenge of federated learning is the presence of heterogeneity. \emph{Data heterogeneity} refers to the fact that data from different clients may follow very different distributions. \emph{System heterogen…

Federated Learning

Initialization Matters: Privacy-Utility Analysis of Overparameterized Neural Networks

2023-10-31 · NeurIPS 2023 11

We analytically investigate how over-parameterization of models in randomized machine learning algorithms impacts the information leakage about their training data. Specifically, we prove a privacy bound for the KL diver…

Where to Begin? From Random to Foundation Model Instructed Initialization in Federated Learning for Medical Image Segmentation

2023-11-27 · Ming Li, Guang Yang

In medical image analysis, Federated Learning (FL) stands out as a key technology that enables privacy-preserved, decentralized data processing, crucial for handling sensitive medical data. Currently, most FL models empl…

Federated LearningImage SegmentationMedical Image AnalysisMedical Image Segmentation+1

Neural network initialization with nonlinear characteristics and information on hierarchical features

2025-11-04 · Hikaru Homma, Jun Ohkubo arxiv

Initialization of neural network parameters, such as weights and biases, has a crucial impact on learning performance; if chosen well, we can even avoid the need for additional training with backpropagation. For example,…