Searcharxiv⌕ Search

arXiv subjects

Xiao Zhang

Publications and source records attributed to Xiao Zhang.

At least 361 records · Page 20Linked to original sources
arXiv2023

P-MMF: Provider Max-min Fairness Re-ranking in Recommender System

In this paper, we address the issue of recommending fairly from the aspect of providers, which has become increasingly essential in multistakeholder recommender systems. Existing studies on provider fairness usually focused on designing proportion fairness (PF) metrics that first consider systematic fairness. However, sociological researches show that to make the market more stable, max-min fairness (MMF) is a better metric. The main reason is that MMF aims to improve the utility of the worst ones preferentially, guiding the system to support the providers in weak market positions. When applying MMF to recommender systems, how to balance user preferences and provider fairness in an online recommendation scenario is still a challenging problem. In this paper, we proposed an online re-ranking model named Provider Max-min Fairness Re-ranking (P-MMF) to tackle the problem. Specifically, P-MMF formulates provider fair recommendation as a resource allocation problem, where the exposure slots are considered the resources to be allocated to providers and the max-min fairness is used as the regularizer during the process. We show that the problem can be further represented as a regularized online optimizing problem and solved efficiently in its dual space. During the online re-ranking phase, a momentum gradient descent method is designed to conduct the dynamic re-ranking. Theoretical analysis showed that the regret of P-MMF can be bounded. Experimental results on four public recommender datasets demonstrated that P-MMF can outperformed the state-of-the-art baselines. Experimental results also show that P-MMF can retain small computationally costs on a corpus with the large number of items.

cs.IR↗
arXiv2023

Semi-Tensor Product of Hypermatrices with Application to Compound Hypermatrices

The semi-tensor product (STP) of matrices is extended to the STP of hypermatrices. Some basic properties of the STP of matrices are extended to the STP of hypermatrices. The hyperdeterminant of hypersquares is introduced. Some algebraic and geometric structures of matrices are extended to hypermatrices. Then the compound hypermatrix is proposed. The STP of hypermatrix is used to compound hypermatrix. Basic properties are proved to be available for compound hypermatrix.

eess.SY↗
arXiv2023

A Study of GeV Gamma-ray Emission toward Supernova Remnant G51.26+0.11 and Its Molecular Environment

We reanalyze the Fermi-LAT GeV $γ$-ray emission in the region of supernova remnant (SNR) G51.26+0.11 and investigate its interstellar molecular environment with the CO-line data. At GeV energies, based on 13.2 years of Fermi-LAT data, the extended $γ$-ray emission observed in this region is resolved into a uniform-disk source ('Src A') with a significance of 19.5$σ$ and a point source (4FGL J1924.3+1628) with a significance of 4.2$σ$ in 0.2$-$500 GeV. With an angular radius of $\sim$ 0.17$^°$, 'Src A' overlaps with SNR G51.26+0.11 significantly in the line of sight. On the other hand, the morphological coincidence between the SNR and the $\sim$ +54 km s$^{-1}$ molecular clouds (MCs) together with the asymmetric or broad $^{12}$CO line profiles near the SNR boundary provides evidence for the very likely SNR-MC interaction. The SNR-MC interaction and the HI absorption features indicate that SNR G51.26+0.11 is located at a kinematic distance of 6.2 $\pm$ 0.5 kpc. Combined with the results from the multi-wavelength analysis, the $γ$-ray emission of the SNR ('Src A') can be naturally explained by a hadronic model with a soft power-law proton spectrum of index $\sim$ 2.25.

astro-ph.HE↗
arXiv2023

Electron-phonon physics from first principles using the EPW code

EPW is an open-source software for $\textit{ab initio}$ calculations of electron-phonon interactions and related materials properties. The code combines density functional perturbation theory and maximally-localized Wannier functions to efficiently compute electron-phonon coupling matrix elements on ultra-fine Brillouin zone grids. This data is employed for predictive calculations of temperature-dependent properties and phonon-assisted quantum processes in bulk solids and low-dimensional materials. Here, we report on significant new developments in the code that occurred during the period 2016-2022, namely: a transport module for the calculation of charge carrier mobility and conductivity under electric and magnetic fields within the $\textit{ab initio}$ Boltzmann transport equation; a superconductivity module for the calculation of critical temperature and gap structure in phonon-mediated superconductors within the $\textit{ab initio}$ anisotropic multi-band Eliashberg theory; an optics module for calculations of phonon-assisted indirect transitions; a module for the calculation of small and large polarons without supercells using the $\textit{ab initio}$ polaron equations; and a module for calculating electron-phonon couplings, band structure renormalization, and temperature-dependent optical spectra using the special displacement method. For each capability, we outline the methodology and implementation, and provide example calculations. We describe recent code refactoring to prepare EPW for exascale architectures, we discuss efficient parallelization strategies, and report on extreme parallel scaling tests.

cond-mat.mtrl-sci↗
arXiv2023

Phonon-assisted optical absorption of SiC polytypes from first principles

Silicon carbide (SiC) is an indirect-gap semiconductor material widely used in electronic and optoelectronic applications. While experimental measurements of the phonon-assisted absorption coefficient of SiC across its indirect gap have existed for more than fifty years, theoretical investigations of phonon-assisted absorption have been hampered by their excessive computational cost. In this work, we calculate the phonon-assisted temperature-dependent optical absorption spectra of the commonly occurring SiC polytypes (3C, 2H, 4H and 6H), using first-principles approaches based on density functional theory and related techniques. We show that our results agree with experimentally determined absorption coefficients in the spectral region between the direct and indirect band gaps. The temperature dependence of the spectra can be well-predicted with taking the temperature-dependence of the band gaps into account. Lastly, we compare the spectra obtained with second-order perturbation theory to those determined by the special displacement method, and we show that the full consideration of the electronic energy renormalization due to temperature is important to further improve the prediction of the phonon-assisted absorption in SiC. Our insights can be applied to predict the optical spectra of the less common SiC polytypes and other indirect-gap semiconductors in general.

cond-mat.mes-hall↗
arXiv2023

Ab initio calculation of carrier mobility in semiconductors including ionized-impurity scattering

The past decade has seen the emergence of ab initio computational methods for calculating phonon-limited carrier mobilities in semiconductors with predictive accuracy. More realistic calculations ought to take into account additional scattering mechanisms such as, for example, impurity and grain-boundary scattering. In this work, we investigate the effect of ionized-impurity scattering on the carrier mobility. We model the impurity potential by a collection of randomly distributed Coulomb scattering centers, and we include this relaxation channel into the ab initio Boltzmann transport equation, as implemented in the EPW code. We demonstrate this methodology by considering silicon, silicon carbide, and gallium phosphide, for which detailed experimental data are available. Our calculations agree reasonably well with experiments over a broad range of temperatures and impurity concentrations. For each compound investigated here, we compare the relative importance of electron-phonon scattering and ionized-impurity scattering, and we critically assess the reliability of Matthiessen's rule. We also show that an accurate description of dielectric screening and carrier effective masses cam improve quantitative agreement with experiments.

cond-mat.mtrl-sci↗
arXiv2022

Bridging The Gap: Entailment Fused-T5 for Open-retrieval Conversational Machine Reading Comprehension

Open-retrieval conversational machine reading comprehension (OCMRC) simulates real-life conversational interaction scenes. Machines are required to make a decision of "Yes/No/Inquire" or generate a follow-up question when the decision is "Inquire" based on retrieved rule texts, user scenario, user question, and dialogue history. Recent studies explored the methods to reduce the information gap between decision-making and question generation and thus improve the performance of generation. However, the information gap still exists because these pipeline structures are still limited in decision-making, span extraction, and question rephrasing three stages. Decision-making and generation are reasoning separately, and the entailment reasoning utilized in decision-making is hard to share through all stages. To tackle the above problem, we proposed a novel one-stage end-to-end framework, called Entailment Fused-T5 (EFT), to bridge the information gap between decision-making and generation in a global understanding manner. The extensive experimental results demonstrate that our proposed framework achieves new state-of-the-art performance on the OR-ShARC benchmark.

cs.CL↗
arXiv2022

Characterizing human collective behaviours of COVID-19 in Hong Kong

People are likely to engage in collective behaviour online during extreme events, such as the COVID-19 crisis, to express their awareness, actions and concerns. Hong Kong has implemented stringent public health and social measures (PHSMs) to curb COVID-19 epidemic waves since the first COVID-19 case was confirmed on 22 January 2020. People are likely to engage in collective behaviour online during extreme events, such as the COVID-19 crisis, to express their awareness, actions and concerns. Here, we offer a framework to evaluate interactions among individuals emotions, perception, and online behaviours in Hong Kong during the first two waves (February to June 2020) and found a strong correlation between online behaviours of Google search and the real-time reproduction numbers. To validate the model output of risk perception, we conducted 10 rounds of cross-sectional telephone surveys from February 1 through June 20 in 2020 to quantify risk perception levels over time. Compared with the survey results, the estimates of the risk perception of individuals using our network-based mechanistic model capture 80% of the trend of people risk perception (individuals who worried about being infected) during the studied period. We may need to reinvigorate the public by engaging people as part of the solution to live their lives with reduced risk.

cs.SI↗
arXiv2022

Ab initio theory of free-carrier absorption in semiconductors

The absorption of light by free carriers in semiconductors results in optical loss for all photon wavelengths. Since free-carrier absorption competes with optical transitions across the band gap, it also reduces the efficiency of optoelectronic devices such as solar cells because it does not generate electron-hole pairs. In this work, we develop a first-principles theory of free-carrier absorption taking into account both single-particle excitations and the collective Drude term, and we demonstrate its application to the case of doped Si. We determine the free-carrier absorption coefficient as a function of carrier concentration and we obtain excellent agreement with experimental data. We identify the dominant processes that contribute to free-carrier absorption at various photon wavelengths, and analyze the results to evaluate the impact of this loss mechanism on the efficiency of Si-based optoelectronic devices.

cond-mat.mtrl-sci↗
arXiv2022

Hybrid topological photonic crystals

Photonic topological phases offering unprecedented manipulation of electromagnetic waves have attracted much research interest which, however, have been mostly restricted to a single band gap. Here, we report on the experimental discovery of hybrid topological photonic crystals which host simultaneously quantum anomalous Hall and valley Hall phases in different photonic band gaps. The underlying hybrid topological phase manifests itself in the edge responses as the coexistence of the chiral edge states and valley Hall edge states in different frequency ranges. We experimentally verify such an emergent phenomenon and show that such a feature enables novel multiplexing of photon transport in the edge channels. Our study reveals a situation with coexisting topology of distinct nature in a single photonic system that may enable frequency-dependent filtering and manipulation of topological edge photons.

cond-mat.mes-hall↗
arXiv2022

PseudoAugment: Learning to Use Unlabeled Data for Data Augmentation in Point Clouds

Data augmentation is an important technique to improve data efficiency and save labeling cost for 3D detection in point clouds. Yet, existing augmentation policies have so far been designed to only utilize labeled data, which limits the data diversity. In this paper, we recognize that pseudo labeling and data augmentation are complementary, thus propose to leverage unlabeled data for data augmentation to enrich the training data. In particular, we design three novel pseudo-label based data augmentation policies (PseudoAugments) to fuse both labeled and pseudo-labeled scenes, including frames (PseudoFrame), objecta (PseudoBBox), and background (PseudoBackground). PseudoAugments outperforms pseudo labeling by mitigating pseudo labeling errors and generating diverse fused training scenes. We demonstrate PseudoAugments generalize across point-based and voxel-based architectures, different model capacity and both KITTI and Waymo Open Dataset. To alleviate the cost of hyperparameter tuning and iterative pseudo labeling, we develop a population-based data augmentation framework for 3D detection, named AutoPseudoAugment. Unlike previous works that perform pseudo-labeling offline, our framework performs PseudoAugments and hyperparameter tuning in one shot to reduce computational cost. Experimental results on the large-scale Waymo Open Dataset show our method outperforms state-of-the-art auto data augmentation method (PPBA) and self-training method (pseudo labeling). In particular, AutoPseudoAugment is about 3X and 2X data efficient on vehicle and pedestrian tasks compared to prior arts. Notably, AutoPseudoAugment nearly matches the full dataset training results, with just 10% of the labeled run segments on the vehicle detection task.

cs.CV↗
arXiv2022

Spin triplet superconducting pairing in doped MoS$_2$

Searching for triplet superconductivity has been pursued intensively in a broad field of material science and quantum information for decades. Nevertheless, these novel states remain rare. Within a simplified effective three-orbital model, we reveal a spin triplet pairing in doped MoS$_2$ by employing both the finite temperature determinant quantum Monte Carlo approach and the ground state constrained-phase quantum Monte Carlo method. In a wide filling region of $\avg{n}=0.60-0.80$ around charge neutrality $\avg{n}=2/3$, the $f$-wave pairing dominates over other symmetries. The pairing susceptibility strongly increases as the on-site Coulomb interaction increases, and it is insensitive to spin-orbit coupling.

cond-mat.supr-con↗
arXiv2022

ET5: A Novel End-to-end Framework for Conversational Machine Reading Comprehension

Conversational machine reading comprehension (CMRC) aims to assist computers to understand an natural language text and thereafter engage in a multi-turn conversation to answer questions related to the text. Existing methods typically require three steps: (1) decision making based on entailment reasoning; (2) span extraction if required by the above decision; (3) question rephrasing based on the extracted span. However, for nearly all these methods, the span extraction and question rephrasing steps cannot fully exploit the fine-grained entailment reasoning information in decision making step because of their relative independence, which will further enlarge the information gap between decision making and question phrasing. Thus, to tackle this problem, we propose a novel end-to-end framework for conversational machine reading comprehension based on shared parameter mechanism, called entailment reasoning T5 (ET5). Despite the lightweight of our proposed framework, experimental results show that the proposed ET5 achieves new state-of-the-art results on the ShARC leaderboard with the BLEU-4 score of 55.2. Our model and code are publicly available at https://github.com/Yottaxx/ET5.

cs.CL↗
arXiv2022

Invariant and Dual Invariant Subspaces of $k$-valued Networks

Consider a $k$-valued network. Two kinds of (control) invariant subspaces, called state and dual invariant subspaces, are proposed, which are subspaces of state space and dual space respectively. Algorithms are presented to verify whether a dual subspace is a dual or dual control invariant subspace. The bearing space of $k$-valued (control) networks is introduced. Using the structure of bearing space, the universal invariant subspace is introduced, which is independent of the dynamics of particular networks. Finally, the relationship between state invariant subspace and dual invariant subspace of a network is investigated. A duality property shows that if a dual subspace is invariant then its perpendicular state subspace is also invariant and vice versa.

eess.SY↗
arXiv2022

MICO: Selective Search with Mutual Information Co-training

In contrast to traditional exhaustive search, selective search first clusters documents into several groups before all the documents are searched exhaustively by a query, to limit the search executed within one group or only a few groups. Selective search is designed to reduce the latency and computation in modern large-scale search systems. In this study, we propose MICO, a Mutual Information CO-training framework for selective search with minimal supervision using the search logs. After training, MICO does not only cluster the documents, but also routes unseen queries to the relevant clusters for efficient retrieval. In our empirical experiments, MICO significantly improves the performance on multiple metrics of selective search and outperforms a number of existing competitive baselines.

cs.IR↗
arXiv2022

Spherically symmetric Einstein-scalar-field equations for wave-like decaying null infinity

We show that the spherically symmetric Einstein-scalar-field equations for wave-like decaying initial data at null infinity have unique local solutions and unique global solutions for small initial data. We also generalize Christodoulou's global generalized solutions to the wave-like decaying initial data. We emphasize that this decaying condition is sharp.

gr-qc↗
arXiv2022

Adversarial Attacks on ASR Systems: An Overview

With the development of hardware and algorithms, ASR(Automatic Speech Recognition) systems evolve a lot. As The models get simpler, the difficulty of development and deployment become easier, ASR systems are getting closer to our life. On the one hand, we often use APPs or APIs of ASR to generate subtitles and record meetings. On the other hand, smart speaker and self-driving car rely on ASR systems to control AIoT devices. In past few years, there are a lot of works on adversarial examples attacks against ASR systems. By adding a small perturbation to the waveforms, the recognition results make a big difference. In this paper, we describe the development of ASR system, different assumptions of attacks, and how to evaluate these attacks. Next, we introduce the current works on adversarial examples attacks from two attack assumptions: white-box attack and black-box attack. Different from other surveys, we pay more attention to which layer they perturb waveforms in ASR system, the relationship between these attacks, and their implementation methods. We focus on the effect of their works.

cs.SD↗