arXiv subjects

Jie Ma

Publications and source records attributed to Jie Ma.

At least 217 records · Page 12Linked to original sources

arXiv2022

Ta4CoSi: a Tantalum-rich superconductor with a honeycomb network structure

The crystal structure of solid-state materials with unique lattices is considered one of the main factors determining physical properties, including superconductivity. Materials with honeycomb lattices have inspired intense research interest for their novel properties. A previously unknown compound, Ta4CoSi, are discussed herein and its crystal structure and superconducting properties. It crystallizes in a CuAl2-type structure (space group P4/mcm, No. 124) with the lattice parameters a = b = 6.17755(7)Å and c = 4.98443(8) Å, featuring honeycomb networks of Ta-Ta in (110) plane. Superconductivity is observed below the critical temperature of 2.45 K, while lower and upper critical magnetic fields are 9.86 mT and 0.84 T, respectively. Experiments and theoretical calculations show that the honeycomb lattices significantly influence the superconducting properties. This material may thus provide a new platform for investigating the exotic superconductivity of honeycomb networks.

cond-mat.supr-con↗

arXiv2022

Federated Learning from Pre-Trained Models: A Contrastive Learning Approach

Federated Learning (FL) is a machine learning paradigm that allows decentralized clients to learn collaboratively without sharing their private data. However, excessive computation and communication demands pose challenges to current FL frameworks, especially when training large-scale models. To prevent these issues from hindering the deployment of FL systems, we propose a lightweight framework where clients jointly learn to fuse the representations generated by multiple fixed pre-trained models rather than training a large-scale model from scratch. This leads us to a more practical FL problem by considering how to capture more client-specific and class-relevant information from the pre-trained models and jointly improve each client's ability to exploit those off-the-shelf models. In this work, we design a Federated Prototype-wise Contrastive Learning (FedPCL) approach which shares knowledge across clients through their class prototypes and builds client-specific representations in a prototype-wise contrastive manner. Sharing prototypes rather than learnable model parameters allows each client to fuse the representations in a personalized way while keeping the shared knowledge in a compact form for efficient communication. We perform a thorough evaluation of the proposed FedPCL in the lightweight framework, measuring and visualizing its ability to fuse various pre-trained models on popular FL datasets.

cs.CR↗

arXiv2022

Significant thermal Hall effect in the 3$d$ cobalt Kitaev system $Na_2Co_2TeO_6$

Kitaev physics has recently attracted attention in condensed matter for its anticipated novel quantum spin liquid state. The thermal transport measurement is crucial for probing the novel features of charge-neutral quasiparticles. In this letter, we report a significant thermal Hall effect in $Na_2Co_2TeO_6$ (NCTO), a Kitaev quantum spin liquid candidate, when the magnetic field is applied along the out-of-plane direction of the honeycomb plane. The thermal conductivity ($κ_{xx}$) and thermal Hall conductivity ($κ_{xy}$) in NCTO reveal distinct magnetic field dependences below and above the Neel temperature ($T_N$) of 27 K. For $T>T_N$, $κ_{xx}$ has a monotonic decrease in the field dependence, while $κ_{xy}$ persists up to $T^*$ = 150 K. On the other hand, both $κ_{xx}$ and $κ_{xy}$ exhibit complex field dependence for $T<T_N$.

cond-mat.str-el↗

arXiv2022

Low-energy Spin Dynamics of Quantum Spin Liquid Candidate $NaYbSe_{2}$

The family of rare earth chalcogenides $ARECh_{2}$ (A = alkali or monovalent ions, RE = rare earth, and Ch = O, S, Se, and Te) appears as an inspiring playground for studying quantum spin liquids (QSL). The crucial low-energy spin dynamics remain to be uncovered. By employing muon spin relaxation ($μ$SR) and zero-field (ZF) AC susceptibility down to 50 mK, we are able to identify the gapless QSL in $NaYbSe_{2}$, a representative member with an effective spin-1/2, and explore its unusual spin dynamics. The ZF $μ$SR experiments unambiguously rule out spin ordering or freezing in $NaYbSe_{2}$ down to 50 mK, two orders of magnitude smaller than the exchange coupling energies. The spin relaxation rate, $λ$, approaches a constant below 0.3 K, indicating finite spin excitations featured by a gapless QSL ground state. This is consistently supported by our AC susceptibility measurements. The careful analysis of the longitudinal field (LF) $μ$SR spectra reveals a strong spatial correlation and a temporal correlation in the spin-disordered ground state, highlighting the unique feature of spin entanglement in the QSL state. The observations allow us to establish an experimental H-T phase diagram. The study offers insight into the rich and exotic magnetism of the rare earth family.

cond-mat.str-el↗

arXiv2022

A non-uniform extension of Baranyai's Theorem

A celebrated theorem of Baranyai states that when $k$ divides $n$, the family $K_n^k$ of all $k$-subsets of an $n$-element set can be partitioned into perfect matchings. In other words, $K_n^k$ is $1$-factorable. In this paper, we determine all $n, k$, such that the family $K_n^{\le k}$ consisting of subsets of $[n]$ of size up to $k$ is $1$-factorable, and thus extend Baranyai's Theorem to the non-uniform setting. In particular, our result implies that for fixed $k$ and sufficiently large $n$, $K_n^{\le k}$ is $1$-factorable if and only if $n \equiv 0$ or $-1 \pmod k$.

math.CO↗

arXiv2022

On the Convergence of Clustered Federated Learning

Knowledge sharing and model personalization are essential components to tackle the non-IID challenge in federated learning (FL). Most existing FL methods focus on two extremes: 1) to learn a shared model to serve all clients with non-IID data, and 2) to learn personalized models for each client, namely personalized FL. There is a trade-off solution, namely clustered FL or cluster-wise personalized FL, which aims to cluster similar clients into one cluster, and then learn a shared model for all clients within a cluster. This paper is to revisit the research of clustered FL by formulating them into a bi-level optimization framework that could unify existing methods. We propose a new theoretical analysis framework to prove the convergence by considering the clusterability among clients. In addition, we embody this framework in an algorithm, named Weighted Clustered Federated Learning (WeCFL). Empirical analysis verifies the theoretical results and demonstrates the effectiveness of the proposed WeCFL under the proposed cluster-wise non-IID settings.

cs.LG↗

arXiv2022

Observation of interaction-induced mobility edge in a disordered atomic wire

Mobility edge, a critical energy separating localized and extended excitations, is a key concept for understanding quantum localization. Aubry-André (AA) model, a paradigm for exploring quantum localization, does not naturally allow mobility edges due to self-duality. Using the momentum-state lattice of quantum gas of Cs atoms to synthesize a nonlinear AA model, we provide experimental evidence for mobility edge induced by interactions. By identifying the extended-to-localized transition of different energy eigenstates, we construct a mobility-edge phase diagram. The location of mobility edge in the low- or high-energy region is tunable via repulsive or attractive interactions. Our observation is in good agreement with the theory, and supports an interpretation of such interaction-induced mobility edge via a generalized AA model. Our work also offers new possibilities to engineer quantum transport and phase transitions in disordered systems.

quant-ph↗

arXiv2022

Regulate the direct-indirect electronic band gap transition by electron-phonon interaction in BaSnO3

The neutron powder diffraction, specific heat, thermal conductivity, and Raman scattering measurements were presented to study the interplays of lattice, phonons and electrons of the Sr-doping Ba1-xSrxSnO3 (x was less than or equal to 0.1). Although Ba1-xSrxSnO3 kept the cubic lattice, the Raman spectra suggested a dynamic distortion at low temperature. The density functional theory was applied to analyze the electronic structures and phonon dispersions of Ba1-xSrxSnO3(x = 0, 0.0125), and the behaviors of electron bands around Fermi levels were discussed. According to the experimental and theoretical results, the Sr-doping played a significant role in tuning the indirect band gap of BaSnO3 and influenced the electron-phonon interaction.

cond-mat.mtrl-sci↗

arXiv2022

Label Semantics for Few Shot Named Entity Recognition

We study the problem of few shot learning for named entity recognition. Specifically, we leverage the semantic information in the names of the labels as a way of giving the model additional signal and enriched priors. We propose a neural architecture that consists of two BERT encoders, one to encode the document and its tokens and another one to encode each of the labels in natural language format. Our model learns to match the representations of named entities computed by the first encoder with label representations computed by the second encoder. The label semantics signal is shown to support improved state-of-the-art results in multiple few shot NER benchmarks and on-par performance in standard benchmarks. Our model is especially effective in low resource settings.

cs.CL↗

arXiv2022

Antiferromagnetic Kitaev interaction in $J_\rm{eff}=1/2$ cobalt honeycomb materials Na$_3$Co$_2$SbO$_6$ and Na$_2$Co$_2$TeO$_6$

Finding new materials with antiferromagnetic (AFM) Kitaev interaction is an urgent issue to broaden and enrich the quantum magnetism research significantly. By carrying out inelastic neutron scattering experiments and subsequent analysis, we conclude that Na$_3$Co$_2$SbO$_6$ and Na$_2$Co$_2$TeO$_6$ are new honeycomb cobalt-based AFM Kitaev systems. The spin-orbit excitons at 20-28~meV in both compounds strongly supports the idea that Co$^{2+}$ ions of both compounds have a spin-orbital entangled $J_\rm{eff}=1/2$ state. Furthermore, we found that a generalized Kitaev-Heisenberg Hamiltonian can well describe the spin-wave excitations of both compounds with additional 3rd nearest-neighbor interaction. Our best-fit parameters show large AFM Kitaev terms and off-diagonal symmetric anisotropy terms of a similar magnitude in both compounds. We should stress that our parameters' optimized magnetic structures are consistent with the magnetic structures reported from neutron diffraction studies. Moreover, there is also the magnon-damping effect at the higher energy part of the spin waves, as usually observed in other Kitaev magnets. We demonstrate that Na$_3$Co$_2$SbO$_6$ and Na$_2$Co$_2$TeO$_6$ are the first experimental realization of AFM Kitaev magnets based on the systematic studies of the spin waves and analysis.

cond-mat.str-el↗

arXiv2022

Visualizing and Understanding Patch Interactions in Vision Transformer

Vision Transformer (ViT) has become a leading tool in various computer vision tasks, owing to its unique self-attention mechanism that learns visual representations explicitly through cross-patch information interactions. Despite having good success, the literature seldom explores the explainability of vision transformer, and there is no clear picture of how the attention mechanism with respect to the correlation across comprehensive patches will impact the performance and what is the further potential. In this work, we propose a novel explainable visualization approach to analyze and interpret the crucial attention interactions among patches for vision transformer. Specifically, we first introduce a quantification indicator to measure the impact of patch interaction and verify such quantification on attention window design and indiscriminative patches removal. Then, we exploit the effective responsive field of each patch in ViT and devise a window-free transformer architecture accordingly. Extensive experiments on ImageNet demonstrate that the exquisitely designed quantitative method is shown able to facilitate ViT model learning, leading the top-1 accuracy by 4.28% at most. Moreover, the results on downstream fine-grained recognition tasks further validate the generalization of our proposal.

cs.CV↗

arXiv2022

Nontrivial topological states in BaSn5 superconductor probed by de Haas-van Alphen quantum oscillations

We report herein the nontrivial topological states in an intrinsic type-II superconductor BaSn5 (Tc ~ 4.4 K) probed via measuring the magnetizations, specific heat, de Haas-van Alphen (dHvA) effect and performing first principles calculations. The first principles calculations reveal a topological nodal ring structure centering at the H point in the kz = π plane of the Brillouin zone (BZ), which could be gapped by spin-orbit coupling (SOC), yielding rather small gaps below and above the Fermi level about 0.04 eV and 0.14 eV, respectively. The SOC also results in a pair of Dirac points along the Γ-A direction and located ~ 0.2 eV above the Fermi level. The analysis of the dHvA quantum oscillations supports the calculations by revealing nontrivial Berry phase originated from three hole and one electron pockets related to the bands forming the Dirac cones. Our study thus provides an excellent avenue for investigating the interplay between superconductivity and nontrivial topological states.

cond-mat.supr-con↗

arXiv2022

Disorder-induced linear magnetoresistance in Sr-doped Bi2Se3 thin films

Sr-doped Bi2Se3 thin films was known as a potential candidate of topological superconductor. The magnetoresistance (MR) of SrxBi2Se3 films with various doping concentrations x were found to be dominated by weak antilocalization (WAL) at low magnetic fields, whereas the classical MR, which originally dominated the MR, was almost completely suppressed. In contrast, the MR of all samples has been observed to be dominated by linear magnetoresistance (LMR) at high magnetic fields. The LMR, having the linear dependence on carrier mobility, can be successfully explained by the Parish-Littlewood model. This indicates that LMR originates from mobility fluctuation induced by Sr dopant atoms in doped Bi2Se3 films.

cond-mat.supr-con↗

arXiv2022

The minimum number of clique-saturating edges

Let $G$ be a $K_p$-free graph. We say $e$ is a $K_p$-saturating edge of $G$ if $e\notin E(G)$ and $G+e$ contains a copy of $K_p$. Denote by $f_p(n, e)$ the minimum number of $K_p$-saturating edges that an $n$-vertex $K_p$-free graph with $e$ edges can have. Erdős and Tuza conjectured that $f_4(n,\lfloor n^2/4\rfloor+1)=\left(1 + o(1)\right)\frac{n^2}{16}.$ Balogh and Liu disproved this by showing $f_4(n,\lfloor n^2/4\rfloor+1)=(1+o(1))\frac{2n^2}{33}$. They believed that a natural generalization of their construction for $K_p$-free graph should also be optimal and made a conjecture that $f_{p+1}(n,ex(n,K_p)+1)=\left(\frac{2(p-2)^2}{p(4p^2-11p+8)}+o(1)\right)n^2$ for all integers $p\ge 3$. The main result of this paper is to confirm the above conjecture of Balogh and Liu.

math.CO↗

arXiv2021

Towards a conjecture of Birmelé-Bondy-Reed on the Erdős-Pósa property of long cycles

A conjecture of Birmelé, Bondy and Reed states that for any integer $\ell\geq 3$, every graph $G$ without two vertex-disjoint cycles of length at least $\ell$ contains a set of at most $\ell$ vertices which meets all cycles of length at least $\ell$. They showed the existence of such a set of at most $2\ell+3$ vertices. This was improved by Meierling, Rautenbach and Sasse to $5\ell/3+29/2$. Here we present a proof showing that at most $3\ell/2+7/2$ vertices suffice.

math.CO↗

arXiv2021

Structured Prediction as Translation between Augmented Natural Languages

We propose a new framework, Translation between Augmented Natural Languages (TANL), to solve many structured prediction language tasks including joint entity and relation extraction, nested named entity recognition, relation classification, semantic role labeling, event extraction, coreference resolution, and dialogue state tracking. Instead of tackling the problem by training task-specific discriminative classifiers, we frame it as a translation task between augmented natural languages, from which the task-relevant information can be easily extracted. Our approach can match or outperform task-specific models on all tasks, and in particular, achieves new state-of-the-art results on joint entity and relation extraction (CoNLL04, ADE, NYT, and ACE2005 datasets), relation classification (FewRel and TACRED), and semantic role labeling (CoNLL-2005 and CoNLL-2012). We accomplish this while using the same architecture and hyperparameters for all tasks and even when training a single model to solve all tasks at the same time (multi-task learning). Finally, we show that our framework can also significantly improve the performance in a low-resource regime, thanks to better use of label semantics.

cs.LG↗

arXiv2021

Tunable intrinsic ferromagnetic topological phases in bulk van der Waals crystal MnSb6Te10

Intrinsic ferromagnetism is a crucial ingredient to realize quantum anomalous Hall effect in quasi two dimensional materials, thus the search of intrinsic ferromagnetic topological materials is one of the most concerned issues in the field of topological phases of matter. In this work, combining magnetotransport measurements, first principles calculations, and angle-resolved photoemission spectroscopy studies, we find that in MnSb6Te10, the n = 2 member of the MnSb2Te4/(Sb2Te3)n family, the strong magnetic competition realizes a fragile ferromagnetic ground state, which whereas easily enters into ferrimagnetic and the Z_2 antiferromagnetic topological insulator phase with warming to higher temperature. Interestingly, the system stays in an inversion-symmetry-protected axion insulator phase in the ferromagnetic ground state as well as in the external magnetic field driven spin-polarized FM phase and can be converted into a Weyl semimetal with multiple Weyl nodes in the valence bands with hole doping, which are manifested by the measured notable intrinsic anomalous Hall effect. Our work thus provides an intrinsic magnetic topological material which is highly tunable into versatile topological phases by temperature, magnetic field, as well as carrier doping.

cond-mat.mtrl-sci↗

arXiv2021

Extremal problems of Erdős, Faudree, Schelp and Simonovits on paths and cycles

For positive integers $n>d\geq k$, let $ϕ(n,d,k)$ denote the least integer $ϕ$ such that every $n$-vertex graph with at least $ϕ$ vertices of degree at least $d$ contains a path on $k+1$ vertices. Many years ago, Erdős, Faudree, Schelp and Simonovits proposed the study of the function $ϕ(n,d,k)$, and conjectured that for any positive integers $n>d\geq k$, it holds that $ϕ(n,d,k)\leq \lfloor\frac{k-1}{2}\rfloor\lfloor\frac{n}{d+1}\rfloor+ε$, where $ε=1$ if $k$ is odd and $ε=2$ otherwise. In this paper we determine the values of the function $ϕ(n,d,k)$ exactly. This confirms the above conjecture of Erdős et al. for all positive integers $k\neq 4$ and in a corrected form for the case $k=4$. Our proof utilizes, among others, a lemma of Erdős et al. \cite{EFSS89}, a theorem of Jackson \cite{J81}, and a (slight) extension of a very recent theorem of Kostochka, Luo and Zirlin \cite{KLZ}, where the latter two results concern maximum cycles in bipartite graphs. Moreover, we construct examples to provide answers to two closely related questions raised by Erdős et al.

math.CO↗