After its establishment, KubeEdge SIG AI provides a large number of resources for its members, including open source data sets such as industrial quality inspection and robot perception, funded incubation projects such as the Linux Foundation and Open Source Summer, online training for cloud-native edge computing, learning materials such as university courses and textbooks, industry standards such as the Cloud Computing Open Source Industry Alliance Standards and national standards, Huawei World Problem Spark Award, KubeEdge Academic Award and academic awards. KubeEdge SIG AI has released two sub-open source projects: the cloud-native edge intelligent service framework KubeEdge-Sedna and the benchmark test suite KubeEdge-Ianvs.
2. KubeEdge-Sedna: Develop diverse services with a standardized unified framework . KubeEdge SIG AI was released in January 2021. Based on the edge-cloud collaboration capabilities provided by KubeEdge, it helps kuwait mobile phone number list developers achieve cross-cloud-edge collaborative training and collaborative reasoning of different types of services under a standardized unified service framework, such as joint reasoning, incremental learning, federated learning, lifelong learning, etc.
KubeEdge-Sedna can enable seamless migration of existing AI applications, thereby reducing costs, improving model performance, and protecting data privacy. In recent years, it has also attracted much attention in cases such as multilateral collaborative re-identification and satellite-to-satellite collaborative satellites.
KubeEdge-Sedna v0.3 released the industry’s first open source feature for edge-cloud collaborative lifelong learning.
3. KubeEdge-Ianvs: Incubate algorithm applications with benchmarks . In July 2022, KubeEdge SIG AI officially released the cloud-native edge intelligence benchmark suite KubeEdge-Ianvs at KubeEdge Submit 2022. It can provide open source data sets, baseline algorithms, test indicators, and stand-alone testing capabilities to help algorithm developers quickly evaluate the performance of collaborative AI algorithms, shortening the PoC time to 1 month and increasing R&D efficiency by 5 times.