Date：2025/01/14

Time：14:00~16:00

Venue：EC2002

Title：Bridging Nanoseconds to Exascale: AI-Driven Innovations in Heterogeneous Computing for Scientific Discovery

Prof. Hsu's talk introduced tools like hls4ml for converting pre-trained ML models into FPGA firmware and the inference-as-a-Service model using Triton Inference Server for large-scale data processing. Attendees gained insights into AI's role in data-driven discovery and learned about advanced techniques for optimizing AI models in demanding scientific applications, from particle physics and gravitational wave detection to neuroscience, all within the context of heterogeneous computing platforms.