Benjamin Nachman
Stanford University, CA·Physics
Particle and nuclear physics experiments record enormously detailed data about how particles react. However, published results are usually squeezed into a few summary numbers or histograms.
- Fields
Physics
Computer Science
- Topics
particle physics
nuclear physics
high energy physics
- Grants
- 2 known grants, through Sep 2027
What they work on
- Develop an exchange format for machine-learning-based unbinned measurements stored as neural networks, and integrate it into an open-source platform that makes such data findable, accessible, interoperable and reusable (FAIR).
- Test the cyberinfrastructure with a prototype pipeline that runs from a first unbinned measurement of electron-proton collision data to a reinterpretation analysis, and connect the tools to common particle, nuclear and astrophysics software frameworks.
- Develop summer-school curricula and online self-guided tutorials on Monte Carlo event generator algorithms for graduate students and junior postdocs in particle and nuclear physics.
- Run a two-year pilot of short-term traineeships in which graduate students and postdoctoral researchers build Monte Carlo models for real particle and nuclear physics problems.
More about the research
This compression throws away information. It makes experiments hard to compare and limits what future scientists can do with archived data long after a detector shuts down. The field also depends on Monte Carlo simulations, computer programs that use random sampling to predict experimental outcomes, yet few students are formally trained to build them. The two projects here address both gaps: one builds software to share richer measurements, and the other trains junior researchers in Monte Carlo methods.
Why it matters
Sharing minimally compressed measurements could let physicists compare experiments and reanalyze archived collider data for years after an experiment ends. It could also open frontier-level analysis to researchers and students without large computing or detector-simulation resources. Training in Monte Carlo methods could prepare junior researchers for simulation-heavy work in physics, machine learning and quantum computing.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Suyog Shrestha
Washington College, MD·Physics
Search ATLAS data from the Large Hadron Collider for new Higgs-like scalar particles that decay into two bottom quarks and two W bosons, in final states with one or two leptons, a decay channel ATLAS has not yet studied.
Jeffrey Potoff
Wayne State University, MI·Chemistry
Speed up py-MCMD, the tool that links the molecular dynamics program NAMD with the Monte Carlo program GOMC, by cutting the delay in each MC/MD cycle through improved memory and disk usage.
G J Peter Elmer
Princeton University, NJ·Physics
Develop new algorithms for reconstruction and triggering, which turn raw detector signals into particles and select which collisions to keep, for the ATLAS and CMS experiments at the High-Luminosity LHC.
The school
Stanford University: 87 labs on PI Pages, $1.6B a year on research, 739 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.