Login

BIB Mitigation with AI Working Meeting

→ US/Central
John Dervan (University of Tennessee), Kiley Kennedy (Princeton University), Ryan Roberts (Unniversity of Chicago, Enrico Fermi Institute)
Description

https://cern.zoom.us/j/69252447084?pwd=bD4hVZx39K7LvQxW012366Jc3b4yUF.1

 

    • 10:00 AM → 10:20 AM
      Update from Chicago 20m
      Speakers: Daniel Abadjiev (University of Chicago), Ryan Roberts (Unniversity of Chicago, Enrico Fermi Institute)
      2026_04_08 siemens BIB rejection update.pdf
    • 10:20 AM → 10:40 AM
      Update from Princeton 20m
      Speakers: Isobel Ojalvo (Princeton University), Kiley Kennedy (Princeton University), Lino Gerlach (Princeton), Mila Bileska (Princeton University)
      • torchlogix is our python library to train LGNs
        • has functionality to write a trained (frozen, discrete) in C & verilog
        • does this layer-by-layer (only works for sequential models)
        • any optimization must happen at later stage
      • proper compilers (transpilers): first build computational graph and then translate
        • allows optimizations before writing out 
        • also allows arbitrary model structures
          • e.g. smartpixels: conv layer on charge profile and then combine with high-level features
      • da4ml is such a library: its tracer overloads numpy backend
        • we are working on integrating torch, such that it works for torchlogix
    • 10:40 AM → 11:00 AM
      Update from Tennessee 20m
      Speakers: John Dervan (University of Tennessee), Jordan Ashley (University of Tennessee, Knoxville)
      • We've contacted Catherine Schuman to revive the SNN for ECAL study—picking up from some preliminary work done by a UTK undergrad
      • Looking into the ASIC logic for stub formation in CMS OT