Search Talks

We discuss how 3+1-dimensional physics including gravity can arise in the IKKT or IIB matrix model.  The model is related to string theory, but it also provides a direct and accessible definition of  3+1-dimensional physics on suitable backgrounds. This leads to a higher-spin extension of gravity on covariant quantum space-time with manifest volume-preserving diffeos, which is UV finite at one loop. 

Zoom Link: https://pitp.zoom.us/j/94588999721?pwd=VTdxQSt3MGhSMkt3UThCVjNRcXJBQT09

The search for pragmatic observables of quantum gravity remains at the forefront of fundamental physics research. A large set of ideas collectively known as the gauge-gravity duality have proven fruitful in tackling this problem. While such a duality is believed to universally govern gravitational theories, its nature in theories of gravity that describe our universe to a good degree of approximation is still little understood.

In this talk I will discuss efforts in formulating a holographic correspondence for gravity in four-dimensional asymptotically flat spacetimes. The proposed dual theory lives on a two-dimensional celestial sphere at infinity and is constrained by a wide range of symmetries. I present recent evidence for this proposal by showing that it arises naturally in a flat space limit of AdS/CFT. I will illustrate this construction with two related examples: the propagation of a particle in a shockwave background and the high-energy scattering of 2 particles.

Zoom Link: https://pitp.zoom.us/j/97426266597?pwd=UmhncFR6NExFVzI2SlJwdE1LeUlIQT09

We examine the validity of the classical approximation of the waterfall phase transition in hybrid inflation from an effective field theory (EFT) point of view. The EFT is constructed by integrating out the waterfall field fluctuations, up to one-loop order in the perturbative expansion. Assuming slow-roll conditions are obeyed, right after the onset of the waterfall phase, we find the backreaction of the waterfall field fluctuations to the evolution of the system can be dominant. In this case the classical approximation is completely spoiled. We derive the necessary constraint that ensures the validity of the EFT.

Zoom Link: https://pitp.zoom.us/j/96209675696?pwd=ZkZ2NGpRVW9vVkdaY1QrWG5GRFltdz09

Topology has become a powerful tool to classify and understand materials. The standard paradigm divides the energy bands into two groups separated by a gap, or with at most some low-dimensional band crossings. Single-gap topological materials include topological insulators, Chern insulators, and topological semimetals. Recent work has extended these ideas to multi-gap systems, in which the energy bands are separated into three or more groups. The resulting phases can be characterized using the Euler class, and exhibit some interesting properties such as non-Abelian charges in the bulk.

In this talk, I will describe our work using first principles calculations to explore topological phases in materials, including both electronic and phononic spectra. In the context of single-gap topology, I will describe the interplay between temperature and topology in electronic systems; while in the context of multi-gap topology, I will present the manipulation of non-Abelian band nodes in phononic systems.

Zoom Link: https://pitp.zoom.us/j/93085720537?pwd=RlhrVFJZU1RmY284Nm5xUXRJakdMZz09

I discuss how exact, non-perturbative results can be obtained for both optical transport and static susceptibilities in “Hertz-Millis” theories of Fermi surfaces coupled to critical bosons. Such models possess a large emergent symmetry and anomaly structure, which we leverage to fix these quantities. In particular, I will show that in the infrared limit, the boson self energy at zero wave vector is a constant independent of frequency, and the real part of the optical conductivity is purely a delta function Drude peak with no other corrections. I will also obtain exact relations between Fermi liquid parameters as the critical point is approached from the disordered phase. 

Zoom Link:  https://pitp.zoom.us/j/93340611986?pwd=cisrZmFxcEVWZVdrT2tMRVZiVTdRQT09

String theory suggests the existence of multiple axion species forming what is known as an “axiverse”. The axions in this model are thought to have logarithmically-distributed masses extending far below 10^{-21} eV, leading to the presence of ultralight axions. The latter have astrophysical de Broglie wavelength and affect the cosmic structures in which they cluster by erasing small-scale features. In contrast to other ultralight or fuzzy dark matter models, the string axiverse allows for a mixed dark sector with a subdominant ultralight component. Trace amounts of ultralight axions have been shown alleviate some observational discrepancies in cosmology such as the missing satellite problem and the Hubble-S8 tensions. Using an effective field theory approach and data from the Baryon Oscillation Spectroscopic Survey, we reach the strongest constraints on the axion relic density for axions with masses below 10^{-25} eV. To study heavier axions, we develop a new algorithm capable of simulating the formation of large-scale structure in the presence of more than one axion species. Making use of this code, we isolate new observational features in the cosmic web which may help us detect the presence of a plenitude of axions with weak lensing surveys.

Zoom Link: https://pitp.zoom.us/j/95089179013?pwd=NTBIeitScFRhYnBBSnlrcnoyVEhydz09

I will discuss the wide - range of new physics that can be probed with nuclear-spin comagnetometers, from EDM measurements to dark matter direct detection experiments.

I will outline the potential for improvement in these measurements, and how improved quantum control could have a significant impact in our sensitivity.  I will present some new dark matter detection results using comagnetometers.

In this talk I will discuss recent advances in Neural-Network parametrisations of pure and mixed quantum states that can be efficiently sampled.  In the first part I will generalise the Neural Density Operator ansatz originally proposed by Torlai and Melko by combining two general ingredients that can be used to construct deep, autoregressive ansatze that automatically enforce positive definitness.  In the second part, instead, I will show that any matrix product state can be exactly represented by a recurrent neural network with a linear memory update. I will then discuss how to generalise this linear RNN architecture to 2D lattices, comparing both approaches to standard DMRG calculations.

Zoom link:  https://pitp.zoom.us/j/98833163484?pwd=bEZTeTB0c1l1QmkrQXdEc3dBQ0dJZz09