The Fast Machine Learning Lab
Fast ML Lab is a research collective of physicists, engineers, and computer scientists interested in deploying machine learning algorithms for unique and challenging scientific applications. Our projects range from real-time, on-detector and low latency machine learning applications to high-throughput heterogeneous computing big data challenges. We are interested in deploying sophisticated machine learning algorithms to advance the exploration of fundamental physics from the world’s biggest colliders to the most intense particle beams to the cosmos.
Shih-Chieh Hsu
Associate Professor of Physics
My research interests include search for physics beyond the Standard Model, and Machine Learning.
Publications
FPGAs-as-a-Service Toolkit (FaaST)
Heterogeneous computing has the potential for significant gains over traditional computing models. This work represents the first open-source FPGAs-as-a-service toolkit.
GPU coprocessors as a service for deep learning inference in high energy physics
Graphics Processing Units can accelerate algorithms to resolve tremendously growing demands for computing in large scientific experiments. We present a comprehensive exploration of the use of GPU-based hardware acceleration for deep learning inference within the data reconstruction workflow of high energy physics.
FPGA-Accelerated Machine Learning Inference as a Service for Particle Physics Computing
Large-scale particle physics experiments face challenging demands for high-throughput computing resources both now and in the future. New heterogeneous computing paradigms on dedicated hardware with increased parallelization, such as Field Programmable Gate Arrays (FPGAs), offer exciting solutions with large potential gains. Using Project Brainwave by Microsoft to accelerate the ResNet-50 image classification model.
