Physics labs race to shrink AI to microsecond speeds
Scientists are building custom hardware and software to run machine learning inference in microseconds—fast enough to filter particle physics data in real time at CERN. The push reflects a broader industry challenge: as experiments generate more data faster, standard AI tools become bottlenecks, driving demand for specialized computing platforms that could reshape how companies process high-velocity sensor data.
Originaltitel: Roadmap on fast machine learning for science
Abstract The need for microsecond speed machine learning (ML) inference for particle physics experiments has emerged in recent years, in particular for the forthcoming upgrades to the experiments at the Large Hadron Collider at CERN. A community has grown around the need to develop the custom hardware platforms and tools required. The material presented in this report is drawn from the latest workshop held by the fast ML for science community and comprises of a collection of perspectives on the status of fast ML in different scientific domains, and the supporting technology.