Speaker
Description
Fast machine learning (ML) has advanced rapidly over the past decade, demonstrating powerful capabilities in particle and nuclear physics experiments. Neutrino experiments are now beginning to adopt these techniques for real-time triggering, data processing, and event reconstruction. However, their requirements vary widely in detector architecture, input dimensionality, accuracy, throughput, and latency. Addressing this diversity requires a broad and adaptable set of algorithms, hardware platforms, and development tools. In this talk, I will review the emerging opportunities for fast ML in neutrino experiments, discuss the key technical challenges, and present strategies for developing efficient solutions tailored to different experimental needs.