arXiv subjects

Yuhao Ma

Publications and source records attributed to Yuhao Ma.

6 recordsLinked to original sources

arXiv2026

Transport and Temperature: Exact spectrum and Drude weight for the one-dimensional infinite-$U$ Hubbard model

Understanding charge transport in strongly correlated systems remains a central challenge in condensed matter physics, while exact analytical results for finite-temperature transport remain rare. Here, we investigate charge transport in the one-dimensional Hubbard model in the infinite-interaction limit. We first construct the exact \emph{and explicit - i.e. beyond Bethe ansatz} energy spectrum and eigenstates. Building on these explicit eigenstates, we derive a closed-form analytical expression for the charge Drude weight at arbitrary temperatures in the hole-dilute limit with a fixed number of doped holes, as well as analytical expressions in the low- and high-temperature limits at fixed hole density. We further show that, at low temperatures, the leading correction to the charge Drude weight is linear in $T$ when the number of holes is fixed, whereas it is quadratic in $T$ when the hole density is fixed. This analytically controlled example highlights the importance of identifying the underlying thermodynamic trajectory when interpreting finite-size numerical studies of charge transport in more general models.

cond-mat.str-el↗

arXiv2024

AeroDiT: Diffusion Transformers for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows

Real-time and accurate prediction of aerodynamic flow fields around airfoils is crucial for flow control and aerodynamic optimization. However, achieving this remains challenging due to the high computational costs and the non-linear nature of flow physics. Traditional Computational Fluid Dynamics (CFD) methods face limitations in balancing computational efficiency and accuracy, hindering their application in real-time scenarios. To address these challenges, this study presents AeroDiT, a novel surrogate model that integrates scalable diffusion models with transformer architectures to address these challenges. Trained on Reynolds-Averaged Navier-Stokes (RANS) simulation data for high Reynolds-number airfoil flows, AeroDiT accurately captures complex flow patterns while enabling real-time predictions. The model demonstrates impressive performance, with mean relative $L_2$ errors of 0.1, 0.025, and 0.050 for pressure $p$ and velocity components $u_x, u_y$, confirming its reliability. To further enhance physical consistency, we incorporate explicit physics-informed losses based on RANS residuals, including mass and momentum conservation constraints. The transformer-based structure allows for real-time predictions within seconds, enabling efficient aerodynamic simulations. This work underscores the potential of generative machine learning techniques to advance computational fluid dynamics, offering potential solutions to challenges in simulating high-fidelity aerodynamic flows.

physics.flu-dyn↗

arXiv2024

A Centralized-Distributed Transfer Model for Cross-Domain Recommendation Based on Multi-Source Heterogeneous Transfer Learning

Cross-domain recommendation (CDR) methods are proposed to tackle the sparsity problem in click through rate (CTR) estimation. Existing CDR methods directly transfer knowledge from the source domains to the target domain and ignore the heterogeneities among domains, including feature dimensional heterogeneity and latent space heterogeneity, which may lead to negative transfer. Besides, most of the existing methods are based on single-source transfer, which cannot simultaneously utilize knowledge from multiple source domains to further improve the model performance in the target domain. In this paper, we propose a centralized-distributed transfer model (CDTM) for CDR based on multi-source heterogeneous transfer learning. To address the issue of feature dimension heterogeneity, we build a dual embedding structure: domain specific embedding (DSE) and global shared embedding (GSE) to model the feature representation in the single domain and the commonalities in the global space,separately. To solve the latent space heterogeneity, the transfer matrix and attention mechanism are used to map and combine DSE and GSE adaptively. Extensive offline and online experiments demonstrate the effectiveness of our model.

cs.LG↗

arXiv2024

Charge Susceptibility and Kubo Response in Hatsugai-Kohmoto-related Models

We study in depth the charge susceptibility for the band Hatsugai-Kohmoto (HK) and orbital (OHK) models. As either of these models describes a Mott insulator, the charge susceptibility takes on the form of a modified density response function with lower and upper Hubbard bands, thereby giving rise to a multi-pole structure. The particle-hole continuum consists of hot spots along the $\omega$ vs $q$ axis arising from inter-band transitions. Such transitions, which are strongly suppressed in non-interacting systems, obtain here because of the non-rigidity of the Hubbard bands. This modified density response function gives rise to a plasmon dispersion that is inversely dependent on the momentum, resulting in an additional contribution to the conventional f-sum rule. This extra contribution originates from a long-range diamagnetic contribution to the current. This results in a non-commutativity of the long-wavelength ($q\rightarrow 0$) and thermodynamic ($L\rightarrow\infty$) limits. When the correct limits are taken, we find that the Kubo response computed with either open or periodic boundary conditions yields identical results that are consistent with the continuity equation contrary to recent claims. We also show that the long wavelength pathology of the current noted previously also plagues the Anderson impurity model interpretation of dynamical mean-field theory (DMFT).

cond-mat.str-el↗

arXiv2023

A mono-atomic orbital-based 1D topological crystalline insulator

Topological crystalline insulators (TCIs) are classified by topological invariants defined with respect to the crystalline symmetries of their gapped bulk. The bulk-boundary correspondence then links the topological properties of the bulk to robust observables on the edges, e.g., the existence of robust edge modes or fractional charge. In one dimension, TCIs protected by reflection symmetry have been realized in a variety of systems where each unit cell has spatially distributed degrees of freedom (SDoF). However, these realizations of TCIs face practical challenges stemming from the sensitivity of the resulting edge modes to variations in edge termination and to the local breaking of the protective spatial symmetries by inhomogeneity. Here we demonstrate topologically protected edge states in a mono-atomic, orbital-based TCI that mitigates both of these issues. By collapsing all SDoF within the unit cell to a singular point in space, we eliminate the ambiguity in unit cell definition and hence remove a prominent source of boundary termination variability. The topological observables are also more tolerant to disorder in the orbital energies. To validate this concept, we experimentally realize a lattice of mechanical resonators where each resonator acts as an "atom" that harbors two key orbital degrees of freedom having opposite reflection parity. Our measurements of this system provide direct visualization of the $sp$-hybridization between orbital modes that leads to a non-trivial band inversion in the bulk. Furthermore, as the spatial width of the resonators is tuned, one can drive a transition between a topological and trivial phase. In the future we expect our approach can be extended to realize orbital-based obstructed atomic insulators and TCIs in higher dimensions.

cond-mat.mes-hall↗

arXiv2020

The Quantum Skin Hall Effect

The skin effect, which is unique to non-Hermitian systems, can generate an extensive number of eigenstates localized near the boundary in an open geometry. Here we propose that in 2D and 3D other quantities besides charge density are susceptible to the skin effect. We show that 2D and 3D models that are a hybrid between topological insulators and skin-effect systems can have a topological skin effect where an extensive number of topological modes, and the corresponding bulk topological invariant, are pinned to the surface. A key example, which we call the quantum skin Hall effect is constructed from layers of Chern insulators and exhibits an extensive Hall conductance and number of chiral modes bound to surfaces normal to the stack of layers. The same procedure is further extended to other symmetry classes to illustrate that a variety of 1D and 2D topological invariants ($\mathbb{Z}$ or $\mathbb{Z}_2)$ are subject to the skin effect. Indeed, we also propose a hybrid 2D system that exhibits an extensive number of topological corner modes and may be more easily realized in meta-material experiments.

cond-mat.mes-hall↗