Federated learning is a privacy-preserving machine learning paradigm to protect the data of clients against privacy breaches. A lot of work on federated learning considers the cross-device setting where the number of clients is large and the data sample size of each client is low. However, this work focuses on cross-silo settings, where clients are few and have large sample sizes. We consider a fully decentralized setting where clients communicate with their immediate time-varying neighbors without the need for a central aggregator prone to congestion and a single point of failure. Our goal is to address stragglers’ delays in cross-silo settings. Existing algorithms designed to overcome stragglers’ delays work with fixed data distributions. They cannot work in real-time settings, such as wireless communication, characterized by time-varying data distributions. Therefore, this paper proposes two online learning algorithms that work with time-varying data and address stragglers’ delays while guaranteeing differential privacy, strong convergence, and communication efficiency. Using the mirror descent technique, the first proposed algorithm addresses the case where the loss gradient is easily computed while the second proposed algorithm addresses the case where the loss gradient is difficult to compute. Simulation results show the performance of the proposed algorithms.
This report describes the hardware for a prototype sensor system and a processing algorithm developed for application with time dependent, reflectance color values obtained using that device. The system utilizes paper supported porphyrin and metalloporphyrin indicators for chemical detection. Here, we explain the event identification algorithm, variations for application of the algorithm in real-time, and potential adjustments to improve sensitivity and selectivity in the output. The described algorithm has been developed with a view toward minimizing the processing power necessary to facilitate its application onboard highly miniaturized sensor devices. The goal is to develop devices that can be utilized as part of a highly distributed chemical sensing network.