Items list
Description
Think Silicon® with very great pleasure attended yesterday the Eighth International Conference on Performance, Safety and Robustness in Complex Systems and Applications (PESARO 2018) which takes place these days, between 22 and 26 of April, at the NOVOTEL Athenes Hotel in Athens, Greece. PESARO 2018 continues a series of events dedicated to fundamentals, techniques and experiments to specify, design, and deploy systems and applications under given constraints on performance, safety and robustness.
During this event, Dr. Georgios Keramidas, Chief Scientific Officer of Think Silicon S.A., chaired the conference’s special track titled “MAIS: Machine Learning Algorithms in Image and Signal Processing”. Machine learning algorithms are currently a hot topic in research and development for academic and industrial teams. Domains like automotive, process automation, etc. reach out for these methods and already deploy systems exploiting machine algorithms, e.g., for image processing and scene analysis. This means, machine learning met the Embedded System domain where almost endless possibilities for the deployment of these algorithms can be found. Trends like the Internet of Things and Cyber-Physical Systems even underline the demand of these novel techniques.
Within this session the participants had the opportunity to hear about the latest trends around machine learning, image and signal processing, reconfigurable hardware, deep learning, automotive safety and cyber-physical systems. Also, during this session Dr. Keramidas gave a very interesting presentation about the main challenges that engineers face in deploying neural network applications in ultra-low power devices, which mainly refers to the work performed within the frame of the GPU-WEAR project funded from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 717850.
For more information about our activities within the Neural Network domain, please do not hesitate to contact us at info(at)think-silicon.com.