Thesis Defense |
Events on Wednesday, May 7th, 2025
- Search for dark matter recoiling from pencil-thin jets using CMS data with machine learning techniques
- Time: 10:00 am - 12:00 pm
- Place: 5280 CH
- Speaker: Abhishikth Mallampalli, Physics PhD student
- Abstract: The Standard Model (SM) of particle physics serves as the foundational framework describing the fundamental particles and forces that govern the behavior of matter and radiation in the universe, excluding gravity. It provides a comprehensive theory of the electromagnetic, weak, and strong nuclear interactions—three of the four fundamental forces of nature. Despite its incredible success in explaining a vast range of experimental phenomena, it is still incomplete and there are several open questions. This thesis attempts to answer some of these open questions in physics today.
Several new physics models predict particles that are expected to leave signatures of missing transverse momentum in collider experiments. One of the primary motivations for such searches is the astrophysical evidence for dark matter, including galactic rotation curves, gravitational lensing, and observations of the cosmic microwave background. Weakly Interacting Massive Particles (WIMPs) are a leading candidate for dark matter, and this thesis explores the parameter space of two WIMP-inspired models, setting stringent limits on their viability. In addition, searches for extra spacetime dimensions, leptoquarks, and quantum blackholes are also performed. Machine learning techniques are used for these searches.
This thesis also presents an algorithm to mitigate beam-induced background in a future muon collider using fast machine learning
- Host: Sridhara Dasu
- IMPLEMENTATION OF QUANTUM ALGORITHMS WITH NEUTRAL ATOM ARRAYS
- Time: 3:00 pm - 5:00 pm
- Place: 5280 CH
- Speaker: Cody Poole, Physics PhD Student
- Abstract: Quantum computers promise to eventually provide significant algorithmic advantage over classical computers for a variety of problems. Executing algorithms on a physical device requires compiling circuits to the native gate set of your device. We are in the Noisy Intermediate Scale Quantum (NISQ) era of quantum computers where circuit execution depths are severely limited by qubit decoherence and gate errors. Large defect-free atom arrays can be produced by initially loading into traps with ~50% success and rearranging the trapped atoms into a desired pattern to enable enhanced data rates for calibrating control operations and running circuits. We present on our implementation of defect-free array generation using the Hungarian matching algorithm and on a partially parallelized rearrangement algorithm. Quantum computers based on a register of neutral Cs atoms have demonstrated significant improvements in gate fidelities in recent years. We present the first implementation of quantum algorithms on an array of neutral Cs atoms. Algorithms executed include Greenberger-Horne-Zeilinger state preparation, Quantum Phase Estimation of the ground state energy of the Hydrogen molecule, the Quantum Approximate Optimization Algorithm (QAOA) applied to the MAXCUT problem, and the Variational Quantum Eigensolver algorithm applied to finding the ground state energy of the Lipkin model. Looking to the future, high performance quantum error correction (QEC) codes will eventually be necessary to run very deep circuits that promise to eventually provide quantum advantage. We present on a qubit allocation scheme and Rydberg gate protocol that would allow for the implementation of a recently characterized class of quantum QEC codes known as Bivariate Bicycle codes on a 2D array of Cs qubits.
- Host: Mark Saffman