Federated learning (FL) is a machine learning paradigm where a shared central model is learned across multiple distributed client devices while the training data remains on edge devices or local clients. Most prior work on federated learning uses Federated Averaging (FedAvg) as an optimization method for training in a synchronized fashion. This involves independent training at multiple edge devices with synchronous aggregation steps. However, the assumptions made by FedAvg are not realistic given the heterogeneity of devices. In particular, the volume and distribution of collected data vary in the training process due to different sampling rates of edge devices. The edge devices themselves also vary in their available communication bandwidth and system configurations, such as memory, processor speed, and power requirements. This leads to vastly different training times as well as model/data transfer times. Furthermore, availability issues at edge devices can lead to a lack of contribution from specific edge devices to the federated model. In this paper, we present an Asynchronous Online Federated Learning (ASOfed) framework, where the edge devices perform online learning with continuous streaming local data and a central server aggregates model parameters from local clients. Our framework updates the central model in an asynchronous manner to tackle the challenges associated with both varying computational loads at heterogeneous edge devices and edge devices that lag behind or dropout. Experiments on three real-world datasets show the effectiveness of ASO-fed on lowering the overall training cost and maintaining good prediction performance.
The large yellow croaker (Pseudosciaena crocea) is an economically important marine fish in China. Inheritance of 22 heterozygous microsatellite loci was examined in normal crossed diploid families and meio-gynogenetic families in P. crocea. Two gynogenetic families were produced via inhibition of the second polar body in eggs fertilized with UV-irradiated sperm. The ratio of gynogenesis was proven to be 100% and 96.9% in the two families, respectively. Of the 22 examined loci, 4 showed a segregation distortion in both control and gynogenetic families. Microsatellite-centromere (M-C) map distances were examined using 18 loci with normal Mendelian segregation. Estimated recombination rates ranged between 0 and 1.0 under the assumption of complete interference. High recombinant frequencies between heterozygous markers and the centromere were found in large yellow croaker, as in other teleosts. The average recombination frequency was 0.586. Ten loci showed high M-C recombination with frequency greater than 0.67. M-C distances provide useful information for gene mapping in large yellow croaker.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.