Neurala, an AI-powered visual inspection solutions provider, recently announced that its Brain Builder SDK has now been optimized to enable edge learning for use cases such as manufacturing. Brain Builder is Neurala’s vision AI platform. In its latest release, Brain Builder allows users to build, deploy and analyze customized AI solutions based on learning from a network edge device.
“Traditional approaches to training deep neural networks (DNN) often fall short in deployment when the network encounters a new situation at the edge that it was not trained to classify,” Massimiliano Versace, Neurala CEO, commented in a press release. “This issue is compounded by the fact that oftentimes, DNNs are trained on a server, and then deployed on a smaller, less powerful compute edge that is incapable of learning new information.”
Brain Builder’s novel approach to machine training allows for continued learning and tweaking even after the original training. This occurs at the edge, and the company foresees its platform as a catalyst for an upcoming explosion in edge-friendly AI applications. The company may see some of these applications in the near future as it will test out the SDK in Europe’s largest IoT hackathon next week in Berlin.
Brain Builder utilizes Neurala’s Lifelong-DNN technology. Once models are trained, users can deploy them directly to an edge device through the SDK. The SDK runs on many operating systems, CPUs, GPUs and NPUs. The benefit of running such learning on the edge is that model training takes hours instead of weeks. Learn more at the Neurala site.
Go to Source
Author: <a href="https://www.programmableweb.com/user/%5Buid%5D">ecarter</a>