Budgeted Online Selection of Candidate IoT Clients to Participate in Federated Learning

Abstract

Machine Learning (ML), and Deep Learning (DL) in particular, play a vital role in providing smart services to the industry. These techniques however suffer from privacy and security concerns since data is collected from clients and then stored and processed at a central location. Federated Learning (FL), an architecture in which model parameters are exchanged instead of client data, has been proposed as a solution to these concerns. Nevertheless, FL trains a global model by communicating with clients over communication rounds, which introduces more traffic on the network and increases the convergence time to the target accuracy. In this work, we solve the problem of optimizing accuracy in stateful FL with a budgeted number of candidate clients by selecting the best candidate clients in terms of test accuracy to participate in the training process. Next, we propose an online stateful FL heuristic to find the best candidate clients. Additionally, we propose an IoT client alarm application that utilizes the proposed heuristic in training a stateful FL global model based on IoT device type classification to alert clients about unauthorized IoT devices in their environment. To test the efficiency of the proposed online heuristic, we conduct several experiments using a real dataset and compare the results against state-of-the-art algorithms. Our results indicate that the proposed heuristic outperforms the online random algorithm with up to 27% gain in accuracy. Additionally, the performance of the proposed online heuristic is comparable to the performance of the best offline algorithm.

Federated Deep Learning for Cyber Security in the Internet of Things: Concepts, Applications, and Experimental Analysis

“…The results of simulation on the SEA dataset (i.e., produced by the AT&T Shannon Lab) demonstrate that the proposed system reaches better accuracy and coherence compared to the conventional systems. To find the best candidate clients and solve the issue of accuracy optimization in federated learning, Mohammed et al [61] introduced an online stateful heuristic based on federated learning combined with an IoT client alarm application, which can be used to notify clients of any unauthorized IoT devices in the IoT environment. The results of simulation on a real data set demonstrates that the suggested system surpasses the online randomized algorithm with up to 27% gain in terms of accuracy.…”

Federated Learning for Energy-balanced Client Selection in Mobile Edge Computing

“…A stateful FL heuristic is studied to schedule Internet-of-Things (IoT) devices to improve target accuracy of the FL [13]. The authors in [14] analyze the convergence rate of biased client selection.…”

“…Constraint (11) defines that the total training time of FL in epoch i has to be shorter than the length of the epoch T max . Constraint (12) guarantees that the total bandwidth of the selected clients should be within the bandwidth capacity B. Constraint (13) indicates that the selected clients enable the model accuracy of the server satisfies the lower bound value, which ranges in accuracy from 0 to 1. Moreover, the constraints of client's transmission power and CPU frequency are given in ( 14) and ( 15) , respectively.…”

“…Otherwise, β i,k is set to 0. The total weight of the knapsack has a lower bound which is equal to the minimum requirement of accuracy constraint (13). In addition, it is note worthy that every item to be put into the backpack must obey the time constraint (11).…”

“…FedAECS algorithm returns once a client can satisfy the accuracy constraint (13) of the optimization. Otherwise, FedAECS continues to traverses the list η i until the client β i,m satisfies the accuracy constraint (13). The corresponding objective function value Φ i (β p i ) will be determined; meanwhile, FedAECS further explore the possible combinations of the client β i,m to find the optimal combinationβ i , the corresponding objective function value Φ i (β i ) will be calculated, finally, the optimal objective function value Φ i (β * i ) will be chosen from Φ i (β p i ) and Φ i (β i ).…”