FedTour: Participatory Federated Learning of Tourism Object Recognition Models with Minimal Parameter Exchanges between User Devices
FedTour: Participatory Federated Learning of Tourism Object Recognition Models with Minimal Parameter Exchanges between User Devices
复制标题
DOI：
10.1109/percomworkshops53856.2022.9767391
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Shusaku Tomita;J. P. Talusan;Yugo Nakamura;H. Suwa;K. Yasumoto
中科院分区：
文献类型：
--
作者：
Shusaku Tomita;J. P. Talusan;Yugo Nakamura;H. Suwa;K. Yasumoto
在本文中，我们提出了FedTour，一种基于联邦学习的旅游目标识别模型训练方法，它利用用户设备之间的短距离直接通信，在有限的更新次数内最大化模型性能。在FedTour中，当两个用户设备在一定范围内时，它们首先交换包括各自模型学习程度（例如识别精度）在内的元数据，然后通过使用由不同精度的多对模型训练的回归器来预测合并模型的精度，从而判断是否可以有效地整合对等模型。一旦认为有效，就交换模型参数，并使用FedAvg（两种用户设备模型的平均权重）更新模型。通过仔细设置是否应用FedAvg的阈值，可以在有限的模型参数交换次数内提高模型性能，从而降低用户设备的功耗。我们利用真实观光区实际用户的手机追踪数据进行了模拟，并评估了在将模型参数交换次数限制在40次的情况下，识别10个物体的CNN模型的精度提高情况。结果表明，FedTour将初始模型的准确率提高了112%，而基于八卦的基线方法的准确率为69%。
In this paper, we propose FedTour, a federated learning-based method for training tourism object recognition models, which utilizes short-distance direct communication between user devices and maximizes the model performance within a limited number of updates. In FedTour, whenever two user devices are within range, they first exchange metadata including the learning degree (e.g., recognition accuracy) of their models, and determine whether it is effective to integrate the peer model by using a regressor trained with various pairs of models with different accuracy to predict the accuracy of the merged model. Once it is deemed effective, the model parameters are exchanged and the model is updated using FedAvg (averaging weights of two models of user devices). By carefully setting the threshold of whether FedAvg is applied or not, model performance is improved within a limited number of model parameter exchanges resulting in lower power consumption of user devices. We conducted a simulation using mobile phone trace data of actual users in a real sightseeing area and evaluated the improvement in accuracy of a CNN model that recognizes 10 objects while limiting the number of model parameter exchanges to only 40. Results show FedTour increased the initial model accuracy by 112%, while the baseline gossip-based method achieved 69%.