@Cornkab

Theoretical Physicist (Allegedly) | MIT CTP P.h.D. Student | don't sleep through dreams that can't come true

 Cambridge, MA
rikabgambhir.com
Joined December 2014
  • Tweets181
  • Following316
  • Followers627
9Media
It turns out optimal transport is easy! The "Spectral EMD" lets you compute optimal transport quickly and exactly in closed form -- incredibly useful for QCD and collider physics! arxiv.org/abs/2410.05379

SPECTER: Efficient Evaluation of the Spectral EMD

The Energy Mover's Distance (EMD) has seen use in collider physics as a metric between events and as a geometric method of defining infrared and collinear safe observables. Recently, the Spectral...

arxiv.org
2
7
1
21
3,213
posting this again from last time! arxiv.org/abs/2302.12266
😡fermion, PhD@angryfermion22 Oct 2024
STEM major: what are you learning in school? kindergartener: shapes and colors STEM major: lmao good luck getting a job with that
1
1
15
1,603
Rikab Gambhir retweeted
John M. Martyn@JohnMMartyn1 Oct 2024
In a new work with @bqp_equals_bpp, we parallelize quantum signal processing! arxiv.org/abs/2409.19043 Our parallel QSP algorithm is tailored to property estimation, and distributes the computation over multiple independent quantum devices.
4
7
1
71
5,612
Rikab Gambhir retweeted
John M. Martyn@JohnMMartyn6 Sep 2024
Can classical randomness accelerate quantum algorithms? Indeed, through randomized compiling! arxiv.org/abs/2409.03744 We introduce stochastic quantum signal processing (QSP) as a randomly-compiled version of QSP, and reduce the cost of QSP-based algorithms by a factor of 1/2.
3
10
61
4,965
Great article about #MachineLearning in #ParticlePhysics Theory: symmetrymagazine.org/article…

Machine learning and theory

Theoretical physicists use machine-learning algorithms to speed up difficult calculations and eliminate untenable theories—but could they transform what it means to make discoveries?

symmetrymagazine.org
1
4
324
New #paper on #arXiv today! "Unifying Simulation and Inference with Normalizing Flows" with Haoxing Du, Vinicius Mikuni, Benjamin Nachman (@BPNachman), Ian Pang, and David Shih arxiv.org/abs/2404.18992 (perfect timing for #EuCAIFCon24 this week :-) ) 1/6

Unifying Simulation and Inference with Normalizing Flows

There have been many applications of deep neural networks to detector calibrations and a growing number of studies that propose deep generative models as automated fast detector simulators. We...

arxiv.org
2
2
11
531
#CMSPaper 1286 studies whether, very rarely, top quarks decay without intermediate W bosons, specifically top to 2 other quarks and 1 charged lepton. It sets the world's strongest constraints, improving our knowledge by three to six orders of magnitude 💪 buff.ly/3xBgfwU
1
3
728
Turns out you actually can get a job studying quantum color and shapes! arxiv.org/abs/2302.12266

SHAPER: Can You Hear the Shape of a Jet?

The identification of interesting substructures within jets is an important tool for searching for new physics and probing the Standard Model at colliders. Many of these substructure tools have...

arxiv.org
😡fermion, PhD@angryfermion25 Apr 2024
STEM major: what are you learning in school? kindergartener: colors and shapes STEM major: lmao good luck getting a job with that
1
1
20
2,869
#CMSPaper 1280 measures the strength of the strong nuclear force from the internal sub-blobs inside jets, the spray of particle debris that quarks and gluons leave when they are created. Yes, jets can measure one of the fundamental physics constants! buff.ly/49wJRsI
7
13
1,419
Rikab Gambhir retweeted
Vanessa@vanessainstem21 Apr 2024
had to sneak flour bakery and cafe into my acknowledgements lol
10
3
86
5,239
#CMSPaper 1279 looks for new undiscovered particles decaying to a b quark and a Higgs boson, a b quark and a Z boson, or a t quark and a W boson. These quark-like objects are common in predictions where more Higgs bosons exist. It is a #null result. buff.ly/49BCTm8
7
1
24
1,481
An accelerator known as a muon collider could revolutionize particle physics—if it can be built. Read that story and more this week in Science: scim.ag/6or
96
81
6
246
169,117
Just wrapped up a fun @AspenPhysics talk on "Machine Learning at Future Colliders" indico.cern.ch/event/1305841… based on arxiv.org/abs/2309.06782 w/ @JoosepPata, Farouk Mokhtar, Mengke Zhang, @ericwulff, David Southwick, @MariaGirone4, @a3d3institute
1
5
33
1,867
Give me a Moment to talk about ML for jet tagging: "Moments of Clarity", a generalization of Deep Sets networks! arxiv.org/abs/2403.08854
2
3
1
32
5,671
🎉🎉Out now: OmniJet-α: The first cross-task foundation model for particle physics (2403.05618) Transfer from generative pre-training to supervised classification improves data efficiency by 2-3 orders of magnitude! Thanks to hard work by @joschka_birk & AnnaH @quunihh @UHHMIN
1
4
1
29
3,138
Cool paper I happened to find on arXiv randomly! "Seeing Double" 👀 with @BPNachman arxiv.org/abs/2402.14067

Seeing Double: Calibrating Two Jets at Once

Jet energy calibration is an important aspect of many measurements and searches at the LHC. Currently, these calibrations are performed on a per-jet basis, i.e. agnostic to the properties of other...

arxiv.org
1
5
22
8,193
A recent passion project of mine has been to communicate a research result through an animated video, in an intuitive and pedagogical format. I’m happy to finally share this video, in which I discuss our recent work on neural-network quantum field states: youtu.be/rrvZDZMii-0
5
37
3
273
27,078
We choose to go to the muon [...] not because it is easy but because it is hard
Muon Collider@MuonCollider7 Dec 2023
THIS. IS. OUR. MUON. SHOT!
2
11
802
Rikab Gambhir retweeted
Muon Collider@MuonCollider7 Dec 2023
THIS. IS. OUR. MUON. SHOT!
4
28
5
102
18,451
Sorry for silence recently, busy hosting #ML4Jets in Hamburg. Thanks for all the seriously excellent developments on AI applied to particle physics (& beyond) Not going to summarize the 133 (!) contributions here, but you can check out the slides: indico.cern.ch/event/1253794
1
5
3
44
7,400
Load more