2019 CAP Lecture Tour

Department of Physics Colloquium
Thursday, 14 February 2019 - 2:45 pm to 3:45 pm
Location
Registration
Registration required: 
No
Cost to attend: 
Free of charge
Event organizer: 
Event language: 

Guest speaker: Prof. Bhavin J. Shastri, Engineering Physics Queen’s University Title: Neuromorphic Silicon Photonics Abstract: Artificial Intelligence (AI) is transforming our lives in the same way as the advent of the Internet and cellular phones. AI is revolutionizing the healthcare industry with complex medical data analysis, actualizing self-driving cars, and beating humans at strategy games such as Go. However, it takes thousands of CPUs and GPUs, and many weeks to train the neural networks in AI hardware. Over the last six years, this compute power has doubled every 3.5 months. Traditional CPUs, GPUs, and neuromorphic (i.e., brain-inspired) electronics such as the IBM TrueNorth and Google TPU will not be powerful enough to train the neural networks of the near future. There is a global race to solve this challenge. My research group is developing the next generation of AI hardware with neuromorphic photonics processors that use light instead of electric signals. By combining the high bandwidth and efficiency of photonic devices with the adaptive, parallelism and complexity similar to the brain, our processors have the potential to be at least ten thousand times faster than state-of-the-art electronic processors while consuming less energy. In this lecture, I will describe our work on building processors on a silicon photonics platform which enables large-scale integration of optical and electronic devices on the same substrate. I will demonstrate how our photonic neural networks can solve coupled ordinary differential equations (such as a Lorenz attractor or a nonlinear optimization problem) a thousand times faster than conventional CPUs. In summary, I will take a look at some of the traditional challenges of photonic information processing, describe the photonic neural-network approaches being developed by our lab and others, and offer a glimpse at the future outlook for this emerging field. Refreshments will be served at 2:15 p.m. (BEFORE the seminar) in room ARC 233