Conferences and supporting programme
Machine Learning for Embedded; a System of Software and Hardware Components
Machine learning for embedded systems require the right combination of hardware and software components for optimized and efficient execution. Various tools are available for training models in the cloud. Sensor and vision libraries such as OpenCV and OpenVX as well as tools for pre-trained model translation, compression, optimization, and deployment are some required capabilities for embedded software. On top of that there needs to be a way to deploy models trained in the cloud through the Edge to a variety of heterogeneous compute devices in standard containers. Machine learning hardware acceleration is becoming the defacto requirement for many embedded devices as well. Software packaging and deployment options are varied. So how to put all of this together in a coherent strategy for deploying machine learning in embedded systems? This talk will describe those steps through each phase of the ML lifecycle for embedded systems.
--- Date: 26.02.2019 Time: 12:00 PM - 12:30 PM Location: Conference Counter NCC Ost