postdoc fellow @iaifi_news | PhD @CornellPhysics ‘24 | @cal ‘18 | interested in representation learning for scientific data, optimizers, and the physics of AI
Very excited about this new paper with Eric and the team! It turns out that a well-orchestrated team of Claude Code agents with access to data (and a Claude Max subscription) can autonomously get most of the way to a real physics measurement in about 4 hours!
AI Agents can now automate scientific analyses that took years in mere hours.
Excited to announce our new pre-print on automating the data analysis process using Artificial Intelligence in High Energy Physics and Particle Physics.
With @softcollinear, Andrzej Novak, Dolores Garcia, and Phil Harris we show that just furnishing context (#JFC) with a thorough methodology is enough for AI agents to autonomously perform a HEP analysis end-to-end, complete with a planning, exploration, execution, statistical analysis, all performed by AI agents. Each step reviewed by more AI agents and looped as long as necessary.
We discuss prospects on how the field should evolve given the increasing power of AI models. Attached in the appendices/linked on GitHub are not only the review-ready analysis notes, but we also publish the full, reproducible codebase for each analysis with tracked history, setting a benchmark for openness and replicability in HEP.
Check out the pre-print: https://t.co/A1U8pv8wUL it’s a fun read and an exciting time to be in #AIforScience!
@MIT@MIT_Physics@AnthropicAI@claudeai@iaifi_news@CERN
#AIRevolution #ClaudeCode #AIAgents
Do you want to know what @CMSExperiment is up to in terms of anomaly detection?
Look no further than our note out now! https://t.co/HuHBf0moGd
While no results on data (yet), it gives a shape of the things to come...
Today at #ML4Jets, we present the performance of ML-based anomaly detection techniques for extracting potential new physics phenomena in a model-agnostic way with @CMSExperiment. Very excited and proud of our team for reaching this milestone! https://t.co/mUlU4OtUda!
#CMSpaper: Search for inelastic dark matter in events with two displaced muons and missing transverse momentum in proton-proton collisions at √s = 13 TeV (arXiv:2305.11649) https://t.co/6g5OsZmB6s #NewPhysics
#CMSpaper soon on arXiv: Search for inelastic dark matter in events with two displaced muons and missing transverse momentum in proton-proton collisions at √s = 13 TeV (CERN-EP-2023-083) https://t.co/P212Q71ADf #NewPhysics
CMS physicists recently completed the first search, at the LHC, for “inelastic dark matter”, looking for muon pairs produced far away from the proton-proton collision point, as if they popped out of thin air. #moriond#cern 👉https://t.co/9Krq07MAUn
cms : online DQM is crucial for catching problems with our detectors as soon as possible
also cms : here’s the buggiest shittiest website we could possibly build
the food culture here in france really is amazing! I’ve just visited the platonic ideal of a supermarket: 250 brands of yogurt, no basic selection of fresh herbs. one wants for nothing in this sprawling utopia, this garden of eden
people love to say that LHC experiments “take photos” of collisions, but the photographers among you are probably wondering “what format?”. well, I have it on good authority that ALICE shoots in HEIC (HEavy Ion Collisions)
I’m at CERN! if any of my 2.5 followers are also here please come speak french at me, fence with baguettes, scoff at things, teach me to drive manual, etc etc
My first article for @ParticleBites on CMS's new measurement of the Higgs width & lifetime! This was a lot of fun to write, and I hope it's at least 50% as fun to read https://t.co/86QBEq5JDa
people say “fermilab is in chicago” which is true at some length scale, but really it’s West Of Chicago™️ which turns out to be the worst place in the world