The CTP Lunch Club meets at 12noon in the CTP Cosman seminar room every Friday (provided that there are sufficient speakers). A light lunch will be provided (usually pizza, however some other options may be explored).
The seminars are designed for graduate students and should be accessible to all students. First year students are particularly encouraged to attend so that they may learn about research being performed in the CTP.
Email notification of the club will be sent to the ctp-all, ctp-postdocs and ctp-students email lists as appropriate. If you wish to speak, or have suggestions about speakers and/or possible workshop topics, please contact the organizers: Sarah Geller, Yu-Chien Huang and Vita Gherardo.
Testing many entangled qubits
Soon we may have quantum computers, but how will we know that they're working correctly? One tool that could help answer this question is self-testing: using a large violation of a Bell inequality to verify entanglement between two spatially separated quantum devices. Remarkably, such tests can certify a specific target state without the need to make any assumptions on the internals of the devices. In this talk I will explain the basic idea behind self-testing and its connection to the problem of rigidity of approximate representations of groups. Time permitting, I'll also discuss recent progress towards more efficient self-tests, inspired by the proof of the PCP theorem from computational complexity theory.
Based on joint works with Thomas Vidick (Caltech) and Matthew Coudron (MIT)
Monte Carlo, Sign Problems, and Real-Time Dynamics on the Lattice
From QCD to many-body quantum systems, Monte Carlo methods have shed light on non-perturbative physics with ever increasing precision over the past few decades. Unfortunately, these methods have a limited range of applicability — they often cannot be straightforwardly applied to systems with a finite chemical potential or systems undergoing real time evolution due to the sign problem. Nevertheless, in the past few years a variety of new algorithms have been proposed and successfully used to circumvent the sign problem in a number of otherwise inaccessible models. In this talk, I will review standard Monte Carlo methods, the sign problem, and Lefschetz Thimble-inspired solutions to the sign problem.
21-cm Cosmology and Dark Matter
In this talk, I will discuss 21-cm cosmology, the recent EDGES measurement of a stronger-than-expected 21-cm absorption trough, and the implications of such measurements on our understanding of dark matter annihilation and decay in the early universe.
Track Assisted Mass
Heavy particles decaying at high boost can be difficult to identify due to the relativistic collimation of their decay products. In order to improve the angular sensitivity of searches for these particles, experimentalists at ATLAS have measured the track assisted mass, a version of jet mass which requires angular information from only charged particles. I will discuss a generalization of this idea, and present resummed calculations of this observable computed using the GFF/track-function formalism.
How curved geometry emerges in hydrodynamics
Finite temperature quantum systems can be described by the imaginary-time formalism, where the path integral of the Euclidean action appears. However, it is only applicable to globally thermalized systems, and we have to generalize the imaginary-time formalism to describe locally thermalized systems, e.g. QGP created in high-energy heavy-ion collisions. In this talk, I will explain how we can describe the locally thermalized quantum systems with the help of curved geometry, and show a modern derivation of nondissipative transport including anomaly-induced transport such as the chiral magnetic effect.
Subleading Power Resummation for Event Shapes
n this blackboard talk, I'll explain how to perform resummation at subleading power for observables relevant for collider physics, like for example event shapes.
• I'll briefly review how to perform resummation at leading power and I'll explain what are the issues in going at subleading powers.
• I'll show how to overcome these issues in the context of a toy example.
• Finally, I'll perform the leading log resummation for an event shapes at subleading power, in particular thrust for H -> gg highlighting all the ingredients that are in general necessary to do such a calculation at subleading power
(Machine) Learning Jet Physics
et substructure plays an important role at the Large Hadron Collider (LHC). Recently, powerful machine learning methods have found use in high energy physics, particularly in jet substructure. In this talk, I will discuss recent progress on machine learning methods designed to tackle persistent challenges in jet physics such as tagging, pileup removal, and learning directly from data. Training on data allows us to avoid relying exclusively on imperfect simulations, potentially providing direct access to probe underlying physics. Machine learning will continue to be an exciting tool to address important questions in high energy physics.
... back to all seminars