Talks

In this talk I will use matrix product states (MPS) to study topological families of gapped ground states in one spatial dimension. To such families I will describe how to associate a gerbe, a mathematical structure which generalizes the line bundle associated to gapped ground states in 0d. Nontriviality of the gerbe represents an obstruction to representing the family of ground states with an MPS tensor that is continuous everywhere over parameter space. I will illustrate these constructions using an exactly solvable topological family which exhibits the key physics in a simple manner.

---

Zoom link https://pitp.zoom.us/j/91497524520?pwd=MkFHSW9PeGlMb2lFOTR3Qmo5clU0dz09

In this presentation, I will discuss two distinct topics, one relating to the foundational aspects of LQG and the other concerning its applicational implications.
Firstly, I will explore the U(1)^3 model of Euclidean Quantum Gravity, which serves as an interesting testing ground for the dynamics problem in LQG. With its analogous constraint structure to full gravity, the U(1)^3 model may hold the key to enhanced quantization techniques.

Secondly, I will delve into the asymptotic symmetries of General Relativity in the Ashtekar-Barbero formulation. New parity conditions for the Ashtekar-Barbero variables will be proposed, which do produce non-trivial supertranslation charges at spatial infinity. This development paves the way for investigating the quantum characteristics of supertranslation charges within the context of LQG.

---

Zoom link https://pitp.zoom.us/j/99532986538?pwd=cnU0VnpJbjU4TSt4MEEzVngxb2wvdz09

The last few years have seen rapid progress in the development of quantum low-density parity-check (LDPC) codes. LDPC codes, defined by their constant weight check operators, can have much better parameters than their topological counterparts like the surface code. In particular, a series of pivotal works culminated in the discovery of asymptotically good LDPC codes--those with essentially optimal rate and distance scalings. These codes allow for the possibility fault-tolerant quantum computation with very low overhead. However, for a code to be used in practice, it is necessary to efficiently identify errors from measurement outcomes to get back into the codespace. In this talk, I will present a linear-time decoder for a family of asymptotically good codes called quantum Tanner codes. Furthermore, I will show that quantum Tanner codes support single-shot decoding, which means that one measurement round suffices to perform reliable quantum error correction, even in the presence of measurement errors. These results can be seen as a step toward making quantum LDPC codes more practical.

---

Zoom link https://pitp.zoom.us/j/94286584094?pwd=Q21IekhHZXI4Qlk4Y1B3MnNobmR6UT09

I will explore subtle aspects of rank-one 4d N=2 supersymmetric QFTs through their low-energy Coulomb-branch physics. The low-energy Lagrangian is famously encoded in "the Seiberg-Witten (SW) curve", which is a one-parameter family of elliptic curves. Here I will explain precisely how "global" aspects of the SQFT such as its spectrum of lines, which cannot be read off from the Lagrangian, are encoded into the SW curve -- more precisely, how they are encoded in its Mordell-Weil group of rational sections. In particular, I will
discuss in detail the difference between the pure SU(2) and the pure SO(3) N=2 SYM theories from this low-energy perspective. I will also comment on the global forms of rank-one 5d SCFTs compactified on a circle.

---

Zoom link https://pitp.zoom.us/j/94967350368?pwd=SHZZUnZNL3ZveFc3NElCaVc5YkY3Zz09

Deep learning (DL) methods have demonstrated great potential for extracting rich non-linear information from cosmological fields, a challenge that traditional summary statistics struggle to address. Most of these DL methods are discriminative models, i.e., they directly learn the posterior constraints of cosmological parameters. In this talk, I will make the argument that learning the field-level likelihood function using generative modeling approaches such as Normalizing Flows usually leads to more effective extraction of cosmological information. This approach also enables anomaly detection to improve the robustness of the analysis. To scale the modeling to high dimensional data and improve its generalization capabilities, we further incorporate physical inductive biases, such as symmetries and multiscale structure, into the architecture of the normalizing flow models. On mock weak lensing maps, I will show that the model leads to significant improvement in constraining power compared to power spectrum and alternative DL models. I will also show that it is able to detect domain shifts between training simulations and test data, such as noise miscalibration and baryonic effect, which, if left unaddressed, could introduce systematic biases in parameter constraints. Finally, I will talk about our ongoing work on applying this model to the field-level cosmic shear analysis for HSC.

---

Zoom link https://pitp.zoom.us/j/97478701784?pwd=UVg4TVp1WFArcXNETG5ITGd0S0NuZz09

It is known that the classical O(N) model in dimension d > 3 at its bulk critical point admits three boundary universality classes: the ordinary, the extraordinary and the special. The extraordinary fixed point corresponds to the bulk transition occurring in the presence of an ordered boundary, while the special fixed point corresponds to a boundary phase transition between the ordinary and the extra-ordinary classes. While the ordinary fixed point survives in d = 3, it is less clear what happens to the extraordinary and special fixed points when d = 3 and N is greater or equal to 2. I'll show that formally treating N as a continuous parameter, there exists a critical value Nc > 2 separating two distinct regimes. For N < Nc the extra-ordinary fixed point survives in d = 3, albeit in a modified form: the long-range boundary order is lost, instead, the order parameter correlation function decays as a power of log r. For N > Nc there is no fixed point with order parameter correlations decaying slower than power law. I'll discuss how these findings compare to recent Monte-Carlo studies of classical and quantum spin models.

---

Zoom link https://pitp.zoom.us/j/97209122334?pwd=UHQ2OXR4bnVZREV0SlJOYXphWjh0QT09

Symmetric mass generation (SMG) is a novel interaction-driven mechanism that generates fermion mass without breaking symmetry, unlike the standard Anderson-Higgs mechanism. SMG can occur in the fermion system without quantum anomalies. In this talk, I will focus on the SMG for the systems with finite fermion density, i.e., the Fermi surface. I will discuss the Fermi surface anomaly and Fermi surface SMG. Lastly, I will talk about its application in the newly found nickelate superconductors, where the superconductivity emerges without a nearby spontaneous symmetry-breaking phase.

---

Zoom link https://pitp.zoom.us/j/92511977879?pwd=MGgyZ0tsZ0hUZDMvZ2wzc3hJVmprZz09

Uncertainty relations are a familiar part of any introductory quantum mechanics course. In this talk, I will summarize how uncertainty relations have been re-interpreted and re-expressed in the language of information theory, leading to connections with the geometry of quantum state space and the limits of computational and information processing efficiency. As two particular examples, I will discuss how uncertainty relations allow one to design information-theoretically optimal measurement protocols for function estimation in networks of quantum sensors and how they enable one to bound the speed at which analog quantum computers can possibly perform optimization tasks. Based primarily on arXiv:2110.07613 and arXiv:2210.15687. 

---

Zoom link https://pitp.zoom.us/j/98258695315?pwd=Q2pEcmg5MGhLWmFlR1FPako0NVFlQT09