Selected article for: "abdominal ct and local global"
Author: Xia, Yingda; Yang, Dong; Li, Wenqi; Myronenko, Andriy; Xu, Daguang; Obinata, Hirofumi; Mori, Hitoshi; An, Peng; Harmon, Stephanie; Turkbey, Evrim; Turkbey, Baris; Wood, Bradford; Patella, Francesca; Stellato, Elvira; Carrafiello, Gianpaolo; Ierardi, Anna; Yuille, Alan; Roth, Holger
Title: Auto-FedAvg: Learnable Federated Averaging for Multi-Institutional Medical Image Segmentation
  • Cord-id: h6svvo77
  • Document date: 2021_4_20
    • ID: h6svvo77
    Snippet: Federated learning (FL) enables collaborative model training while preserving each participant's privacy, which is particularly beneficial to the medical field. FedAvg is a standard algorithm that uses fixed weights, often originating from the dataset sizes at each client, to aggregate the distributed learned models on a server during the FL process. However, non-identical data distribution across clients, known as the non-i.i.d problem in FL, could make this assumption for setting fixed aggrega
    Document: Federated learning (FL) enables collaborative model training while preserving each participant's privacy, which is particularly beneficial to the medical field. FedAvg is a standard algorithm that uses fixed weights, often originating from the dataset sizes at each client, to aggregate the distributed learned models on a server during the FL process. However, non-identical data distribution across clients, known as the non-i.i.d problem in FL, could make this assumption for setting fixed aggregation weights sub-optimal. In this work, we design a new data-driven approach, namely Auto-FedAvg, where aggregation weights are dynamically adjusted, depending on data distributions across data silos and the current training progress of the models. We disentangle the parameter set into two parts, local model parameters and global aggregation parameters, and update them iteratively with a communication-efficient algorithm. We first show the validity of our approach by outperforming state-of-the-art FL methods for image recognition on a heterogeneous data split of CIFAR-10. Furthermore, we demonstrate our algorithm's effectiveness on two multi-institutional medical image analysis tasks, i.e., COVID-19 lesion segmentation in chest CT and pancreas segmentation in abdominal CT.

    Search related documents:
    Co phrase search for related documents
    • local model performance and machine learning paradigm: 1
    • local model train and machine learning: 1, 2, 3
    • local performance and loss function: 1
    • local performance and machine learning: 1, 2, 3, 4
    • local performance and machine learning paradigm: 1
    • local training and machine learning: 1, 2, 3
    • locally model train and machine learning: 1
    • locally model train and machine learning paradigm: 1
    • locally train and machine learning: 1
    • locally train and machine learning paradigm: 1
    • locally train model and machine learning: 1
    • locally train model and machine learning paradigm: 1
    • loss function and machine learning: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23
    • loss function and machine learning paradigm: 1
    • loss function give and machine learning: 1