Soheil Ghiasi received his PhD in Computer Science from UCLA in 2004 and received the Harry M. Showman prize from the UCLA College of Engineering in the same year. He has been on the faculty at Department of Electrical and Computer Engineering, UC Davis since 2004, where he directs Laboratory for Embedded and Programmable Systems (LEPS).
I am interested in design methodologies for embedded and cyber-physical systems (CPS), which utilize computing to monitor, service and control various application-specific processes, including those in the physical world. My work deals with modeling, analysis, synthesis and optimization of embedded software, programmable execution platforms (e.g., processors, GPUs and FPGAs) and tools for automating the design process. The area offers an interesting blend of theory and practice: real-world applications give rise to research problems, for which solutions are developed using a combination of analytical and experimental techniques.
Currently, my students and I are working on efficient execution of trained deep convolutional neural networks (an important technique in machine learning), and embedded systems for selected health applications, such as lung function evaluation and transabdominal fetal oximetry.
Energy Efficient Inference using Deep Convolutional Neural Networks (DCNN)
Kino: Monitoring and Mentoring of Physical Exercises
-Design and programming methodologies for parallel and reconfigurable embedded computing platforms
-Innovative applications of digital computing systems in sectors such as medicine, agriculture and machine intelligence