Sort:
Regular Paper Issue
Age-of-Information-Aware Federated Learning
Journal of Computer Science and Technology 2024, 39(3): 637-653
Published: 22 July 2024
Abstract Collect

Federated learning (FL) is an emerging privacy-preserving distributed computing paradigm, enabling numerous clients to collaboratively train machine learning models without the necessity of transmitting clients’ private datasets to the central server. Unlike most existing research where the local datasets of clients are assumed to be unchanged over time throughout the whole FL process, our study addresses such scenarios in this paper where clients’ datasets need to be updated periodically, and the server can incentivize clients to employ as fresh as possible datasets for local model training. Our primary objective is to design a client selection strategy to minimize the loss of the global model for FL loss within a constrained budget. To this end, we introduce the concept of ‘‘Age of Information’’ (AoI) to quantitatively assess the freshness of local datasets and conduct a theoretical analysis of the convergence bound in our AoI-aware FL system. Based on the convergence bound, we further formulate our problem as a restless multi-armed bandit (RMAB) problem. Next, we relax the RMAB problem and apply the Lagrangian Dual approach to decouple it into multiple subproblems. Finally, we propose a Whittle’s Index Based Client Selection (WICS) algorithm to determine the set of selected clients. In addition, comprehensive simulations substantiate that the proposed algorithm can effectively reduce training loss and enhance the learning accuracy compared with some state-of-the-art methods.

Regular Paper Issue
Cloaking Region Based Passenger Privacy Protection in Ride-Hailing Systems
Journal of Computer Science and Technology 2020, 35(3): 629-646
Published: 29 May 2020
Abstract Collect

With the quick development of the sharing economy, ride-hailing services have been increasingly popular worldwide. Although the service provides convenience for users, one concern from the public is whether the location privacy of passengers would be protected. Service providers (SPs) such as Didi and Uber need to acquire passenger and driver locations before they could successfully dispatch passenger orders. To protect passengers’ privacy based on their requirements, we propose a cloaking region based order dispatch scheme. In our scheme, a passenger sends the SP a cloaking region in which his/her actual location is not distinguishable. The trade-off of the enhanced privacy is the loss of social welfare, i.e., the increase in the overall pick-up distance. To optimize our scheme, we propose to maximize the social welfare under passengers’ privacy requirements. We investigate a bipartite matching based approach. A theoretical bound on the matching performance under specific privacy requirements is shown. Besides passengers’ privacy, we allow drivers to set up their maximum pick-up distance in our extended scheme. The extended scheme could be applied when the number of drivers exceeds the number of passengers. Nevertheless, the global matching based scheme does not consider the interest of each individual passenger. The passengers with low privacy requirements may be matched with drivers far from them. To this end, a pricing scheme including three strategies is proposed to make up for the individual loss by allocating discounts on their riding fares. Extensive experiments on both real-world and synthetic datasets show the efficiency of our scheme.

Total 2
1/11GOpage