Detecting Dark Matter with Planets and Stars
Abstract
In this talk, I will discuss how planets and stars can be utilized as novel and powerful detectors of dark matter. I will begin with planets, focusing on how our own Earth can efficiently probe a rare species of dark matter (dark matter sub-components) with a very short collision length, for which conventional underground direct-detection experiments lose sensitivity. I will then turn to stars, highlighting how nuclear transitions in stellar interiors can give rise to ultralight axions, opening distinctive observational windows. Taken together, these examples demonstrate how planetary and stellar environments offer complementary and cost-effective avenues to search for dark matter, extending sensitivity into regions of parameter space that remain largely inaccessible otherwise.
Chasing Dark Matter: From Light to Heavy
Abstract
The nature of dark matter remains one of the key problems in modern physics, as its mass and interaction strengths are uncertain across an enormous range of scales. In this talk, I will present a scale-agnostic perspective and survey strategies to search for dark matter across the full mass spectrum, from ultralight to the heaviest candidates. I will emphasize how ideas and techniques from astrophysics, particle physics, and nuclear physics together can provide a powerful and unified framework for unraveling this mystery. By weaving together insights from these traditionally distinct approaches, I will argue that meaningful progress on dark matter will emerge from the interplay of ideas across traditional boundaries.
An implementation of GKZ hypergeometric series of Feynman integrals inside ‘Regions’
Abstract
We present an implementation of Gel’fand–Kapranov–Zelevinsky (GKZ) hypergeometric series for Feynman integrals within the Method of Regions (MoR). Our approach establishes a direct link between the geometry of Newton polytopes and the analytic asymptotic expansion of Feynman integrals in D dimension. By embedding GKZ solutions associated with distinct regions, we construct an automated framework that identifies contributing sectors, generates the corresponding GKZ-series solutions, and performs their systematic expansion in the dimension regulator. We have developed a Mathematica package, GKZRegions, which can be used to derive boundary conditions for differential equations of master integrals and to evaluate region contributions in effective field theories such as SCET.
Superconducting order parameter manifested by Quasicrystal
Abstract
Recent discovery of the superconducting ground state in quasicrystals (QCs) has opened up an exciting new avenue for superconductivity based on QCs. In this work, we explore the scope by theoretically investigating the behavior of the superconducting order parameter (OP) in various QCs based on the attractive Hubbard model. By systematically analyzing models generated through various growth rules, we elucidate the influence of aperiodicity on the OP amplitudes and how it evolves towards the periodic limit with the change in the structural pattern. We study the evolution of the OP with respect to the temperature, strength of the interaction, and nearest-neighbor hopping amplitude. Our numerical analysis identifies the most favorable QCs and parameter regime that support enhanced onsite pairing amplitudes. Additionally, we provide a comparative analysis of the superconducting transition temperatures across the range of aperiodic configurations. To gain further insight into these systems, we compute thermodynamic quantities which enable us to determine which QCs are most conducive to Cooper pair formation.
THE QUANTUM HALL EFFECT: APPLIED SCIENCE’S GIFT TO FUNDAMENTAL SCIENCE
Abstract
The discoveries of the integer and fractional quantum Hall effect (IQHE/FQHE) in
a two-dimensional system of electrons in 1980-82 gave rise to the field of topological
matter. These discoveries brought about a revolution in our understanding of the
physics of condensed matter and resulted in six scientists winning the Nobel Prize in
Physics.
Since the original discovery, the two-dimensional electron system subjected to a
large perpendicular magnetic field has become a veritable goldmine for the condensed
matter physicist. Completely unexpected and exotic phenomena such as
fractionalization of electronic charge, composite particles, Abelian and non-Abelian
quantum states, and topological phases have entered the lexicon of quantum
condensed matter. In addition, the system has provided examples of previously known
phenomena such as topological spin excitations and charge density wave phases, the
latter coming in several incarnations such as electron crystals, bubble crystals, and
striped phases.
Behind this amazing discovery lies the story of painstaking development of
semiconductor heterostructures, a field motivated in large part by its technological
importance. This was but one in a series of success stories involving the vast field of
semiconductor technology, starting with the original discovery of the transistor by
physicists at Bell Laboratories in 1947. Without these developments in applied research,
the IQHE and FQHE would scarcely have been discovered, and our understanding of
the richness of nature would have been that much poorer.
This talk will trace this interplay of science and technology in the context of
semiconductor heterostructures, culminating in the discovery of the FQHE, and
unfolding of its complex, hierarchical structure. Following that, I will give an overview of
the fascinating phenomena that QHE exemplifies. Finally, I will briefly discuss a couple
of examples from the past decade that my group has been involved in, using a
combination of analytical and state-of-the-art computational techniques. One will
examine a phase in detail, and another the promise of new material platforms.
Simulation-Based Inference
Abstract
Many modern experiments produce extremely high-dimensional data, while relying on complex simulations for their statistical modeling. In such settings, the likelihood is often not available in closed form. Simulation-Based Inference (SBI) is a family of emerging techniques that allow statistical inference on parameters of the statistical model directly using high-dimensional data, even when the likelihoods are not analytically tractable. This is done by leveraging deep-learning techniques to directly build likelihood-based or posterior-based inference models without any intermediate steps that lose information. These methods can be significantly more powerful than traditional approaches that first compress high-dimensional data into low-dimensional summary observable and then build approximate likelihoods as a function of the summary observable. Such compression can reduce sensitivity by discarding how the high-dimensional information varies as a function of the model parameters.
In this talk, I will review the basics of SBI, focusing primarily on applications in ATLAS and CMS analyses at the Large Hadron Collider (LHC), and briefly highlighting applications in other domains. I will also introduce novel developments that allow for scalable inference when the likelihood model depends on hundreds of nuisance parameters describing systematic uncertainties, as is typical in LHC analysis. These developments resulted in first-ever measurements at the LHC using SBI [1, 2].
1. Measurement of off-shell Higgs boson production in the H->ZZ->4l decay channel using a neural simulation-based inference technique in 13 TeV pp collisions with the ATLAS detector
2. An implementation of neural simulation-based inference for parameter estimation in ATLAS