arXiv Science⌕ Search

arXiv subjects

Xiao Luo

Publications and source records attributed to Xiao Luo.

At least 163 records · Page 9Linked to original sources
arXiv2022

Electrically Driven Thermal Infrared Metasurface with Narrowband Emission

Metasurfaces consisting of an array of planar sub-wavelength structures have shown great potentials in controlling thermal infrared radiation, including intensity, coherence, and polarization. These capabilities together with the two-dimensional nature make thermal metasurfaces an ultracompact multifunctional platform for infrared light manipulation. Integrating the functionalities, such as amplitude, phase (spectrum and directionality), and polarization, on a single metasurface offers fascinating device responses. However, it remains a significant challenge to concurrently optimize the optical, electrical, and thermal responses of a thermal metasurface in a small footprint. In this work, we develop a center-contacted electrode line design for a thermal infrared metasurface based on a gold nanorod array, which allows local Joule heating to electrically excite the emission without undermining the localized surface plasmonic resonance. The narrowband emission of thermal metasurfaces and their robustness against temperature nonuniformity demonstrated in this work have important implications for the applications in infrared imaging, sensing, and energy harvesting.

physics.optics↗
arXiv2022

Efficient textual explanations for complex road and traffic scenarios based on semantic segmentation

The complex driving environment brings great challenges to the visual perception of autonomous vehicles. It's essential to extract clear and explainable information from the complex road and traffic scenarios and offer clues to decision and control. However, the previous scene explanation had been implemented as a separate model. The black box model makes it difficult to interpret the driving environment. It cannot detect comprehensive textual information and requires a high computational load and time consumption. Thus, this study proposed a comprehensive and efficient textual explanation model. From 336k video frames of the driving environment, critical images of complex road and traffic scenarios were selected into a dataset. Through transfer learning, this study established an accurate and efficient segmentation model to obtain the critical traffic elements in the environment. Based on the XGBoost algorithm, a comprehensive model was developed. The model provided textual information about states of traffic elements, the motion of conflict objects, and scenario complexity. The approach was verified on the real-world road. It improved the perception accuracy of critical traffic elements to 78.8%. The time consumption reached 13 minutes for each epoch, which was 11.5 times more efficient than the pre-trained network. The textual information analyzed from the model was also accordant with reality. The findings offer clear and explainable information about the complex driving environment, which lays a foundation for subsequent decision and control. It can improve the visual perception ability and enrich the prior knowledge and judgments of complex traffic situations.

cs.CV↗
arXiv2022

A Survey on Deep Hashing Methods

Nearest neighbor search aims to obtain the samples in the database with the smallest distances from them to the queries, which is a basic task in a range of fields, including computer vision and data mining. Hashing is one of the most widely used methods for its computational and storage efficiency. With the development of deep learning, deep hashing methods show more advantages than traditional methods. In this survey, we detailedly investigate current deep hashing algorithms including deep supervised hashing and deep unsupervised hashing. Specifically, we categorize deep supervised hashing methods into pairwise methods, ranking-based methods, pointwise methods as well as quantization according to how measuring the similarities of the learned hash codes. Moreover, deep unsupervised hashing is categorized into similarity reconstruction-based methods, pseudo-label-based methods and prediction-free self-supervised learning-based methods based on their semantic learning manners. We also introduce three related important topics including semi-supervised deep hashing, domain adaption deep hashing and multi-modal deep hashing. Meanwhile, we present some commonly used public datasets and the scheme to measure the performance of deep hashing algorithms. Finally, we discuss some potential research directions in conclusion.

cs.CV↗
arXiv2022

Informative Causality Extraction from Medical Literature via Dependency-tree based Patterns

Extracting cause-effect entities from medical literature is an important task in medical information retrieval. A solution for solving this task can be used for compilation of various causality relations, such as, causality between disease and symptoms, between medications and side effects, between genes and diseases, etc. Existing solutions for extracting cause-effect entities work well for sentences where the cause and the effect phrases are name entities, single-word nouns, or noun phrases consisting of two to three words. Unfortunately, in medical literature, cause and effect phrases in a sentence are not simply nouns or noun phrases, rather they are complex phrases consisting of several words, and existing methods fail to correctly extract the cause and effect entities in such sentences. Partial extraction of cause and effect entities conveys poor quality, non informative, and often, contradictory facts, comparing to the one intended in the given sentence. In this work, we solve this problem by designing an unsupervised method for cause and effect phrase extraction, PatternCausality, which is specifically suitable for the medical literature. Our proposed approach first uses a collection of cause-effect dependency patterns as template to extract head words of cause and effect phrases and then it uses a novel phrase extraction method to obtain complete and meaningful cause and effect phrases from a sentence. Experiments on a cause-effect dataset built from sentences from PubMed articles show that for extracting cause and effect entities, PatternCausality is substantially better than the existing methods with an order of magnitude improvement in the F-score metric over the best of the existing methods.

cs.CL↗
arXiv2021

ARGO: Modeling Heterogeneity in E-commerce Recommendation

Nowadays, E-commerce is increasingly integrated into our daily lives. Meanwhile, shopping process has also changed incrementally from one behavior (purchase) to multiple behaviors (such as view, carting and purchase). Therefore, utilizing interaction data of auxiliary behavior data draws a lot of attention in the E-commerce recommender systems. However, all existing models ignore two kinds of intrinsic heterogeneity which are helpful to capture the difference of user preferences and the difference of item attributes. First (intra-heterogeneity), each user has multiple social identities with otherness, and these different identities can result in quite different interaction preferences. Second (inter-heterogeneity), each item can transfer an item-specific percentage of score from low-level behavior to high-level behavior for the gradual relationship among multiple behaviors. Thus, the lack of consideration of these heterogeneities damages recommendation rank performance. To model the above heterogeneities, we propose a novel method named intra- and inter-heterogeneity recommendation model (ARGO). Specifically, we embed each user into multiple vectors representing the user's identities, and the maximum of identity scores indicates the interaction preference. Besides, we regard the item-specific transition percentage as trainable transition probability between different behaviors. Extensive experiments on two real-world datasets show that ARGO performs much better than the state-of-the-art in multi-behavior scenarios.

cs.IR↗
arXiv2021

CIMON: Towards High-quality Hash Codes

Recently, hashing is widely used in approximate nearest neighbor search for its storage and computational efficiency. Most of the unsupervised hashing methods learn to map images into semantic similarity-preserving hash codes by constructing local semantic similarity structure from the pre-trained model as the guiding information, i.e., treating each point pair similar if their distance is small in feature space. However, due to the inefficient representation ability of the pre-trained model, many false positives and negatives in local semantic similarity will be introduced and lead to error propagation during the hash code learning. Moreover, few of the methods consider the robustness of models, which will cause instability of hash codes to disturbance. In this paper, we propose a new method named {\textbf{C}}omprehensive s{\textbf{I}}milarity {\textbf{M}}ining and c{\textbf{O}}nsistency lear{\textbf{N}}ing (CIMON). First, we use global refinement and similarity statistical distribution to obtain reliable and smooth guidance. Second, both semantic and contrastive consistency learning are introduced to derive both disturb-invariant and discriminative hash codes. Extensive experiments on several benchmark datasets show that the proposed method outperforms a wide range of state-of-the-art methods in both retrieval performance and robustness.

cs.CV↗
arXiv2021

Normalized solutions for Schrödinger system with quadratic and cubic interactions

In this paper, we give a complete study on the existence and non-existence of normalized solutions for Schrödinger system with quadratic and cubic interactions. In the one dimension case, the energy functional is bounded from below on the product of $L^2$-spheres, normalized ground states exist and are obtained as global minimizers. When $N=2$, the energy functional is not always bounded on the product of $L^2$-spheres. We give a classification of the existence and nonexistence of global minimizers. Then under suitable conditions on $b_1$ and $b_2$, we prove the existence of normalized solutions. When $N=3$, the energy functional is always unbounded on the product of $L^2$-spheres. We show that under suitable conditions on $b_1$ and $b_2$, at least two normalized solutions exist, one is a ground state and the other is an excited state. Furthermore, by refining the upper bound of the ground state energy, we provide a precise mass collapse behavior of the ground state and a precise limit behavior of the excited state as $β\rightarrow 0$. Finally, we deal with the high dimensional cases $N\geq 4$. Several non-existence results are obtained if $β<0$. When $N=4$, $β>0$, the system is a mass-energy double critical problem, we obtain the existence of a normalized ground state and its synchronized mass collapse behavior. Comparing with the well studied homogeneous case $β=0$, our main results indicate that the quadratic interaction term not only enriches the set of solutions to the above Schrödinger system but also leads to a stabilization of the related evolution system.

math.AP↗
arXiv2021

Positive normalized solutions to nonlinear elliptic systems in $\R^4$ with critical Sobolev exponent

In this paper, we consider the existence and asymptotic behavior on mass of the positive solutions to the following system: \begin{equation}\label{eqA0.1}\nonumber \begin{cases} -Δu+λ_1u=μ_1u^3+α_1|u|^{p-2}u+βv^2u\quad&\hbox{in}~\R^4,\\ -Δv+λ_2v=μ_2v^3+α_2|v|^{p-2}v+βu^2v\quad&\hbox{in}~\R^4,\\ \end{cases} \end{equation} under the mass constraint $$\int_{\R^4}u^2=a_1^2\quad\text{and}\quad\int_{\R^4}v^2=a_2^2,$$ where $a_1,a_2$ are prescribed, $μ_1,μ_2,β>0$; $α_1,α_2\in \R$, $p\!\in\! (2,4)$ and $λ_1,λ_2\!\in\!\R$ appear as Lagrange multipliers. Firstly, we establish a non-existence result for the repulsive interaction case, i.e., $α_i<0(i=1,2)$. Then turning to the case of $α_i>0 (i=1,2)$, if $2<p<3$, we show that the problem admits a ground state and an excited state, which are characterized respectively by a local minimizer and a mountain-pass critical point of the corresponding energy functional. Moreover, we give a precise asymptotic behavior of these two solutions as $(a_1,a_2)\to (0,0)$ and $a_1\sim a_2$. This seems to be the first contribution regarding the multiplicity as well as the synchronized mass collapse behavior of the normalized solutions to Schrödinger systems with Sobolev critical exponent. When $3\leq p<4$, we prove an existence as well as non-existence ($p=3$) results of the ground states, which are characterized by constrained mountain-pass critical points of the corresponding energy functional. Furthermore, precise asymptotic behaviors of the ground states are obtained when the masses of whose two components vanish and cluster to a upper bound (or infinity), respectively.

math.AP↗
arXiv2021

DNA-GCN: Graph convolutional networks for predicting DNA-protein binding

Predicting DNA-protein binding is an important and classic problem in bioinformatics. Convolutional neural networks have outperformed conventional methods in modeling the sequence specificity of DNA-protein binding. However, none of the studies has utilized graph convolutional networks for motif inference. In this work, we propose to use graph convolutional networks for motif inference. We build a sequence k-mer graph for the whole dataset based on k-mer co-occurrence and k-mer sequence relationship and then learn DNA Graph Convolutional Network (DNA-GCN) for the whole dataset. Our DNA-GCN is initialized with a one-hot representation for all nodes, and it then jointly learns the embeddings for both k-mers and sequences, as supervised by the known labels of sequences. We evaluate our model on 50 datasets from ENCODE. DNA-GCN shows its competitive performance compared with the baseline model. Besides, we analyze our model and design several different architectures to help fit different datasets.

q-bio.GN↗
arXiv2021

Deep Unsupervised Hashing by Distilled Smooth Guidance

Hashing has been widely used in approximate nearest neighbor search for its storage and computational efficiency. Deep supervised hashing methods are not widely used because of the lack of labeled data, especially when the domain is transferred. Meanwhile, unsupervised deep hashing models can hardly achieve satisfactory performance due to the lack of reliable similarity signals. To tackle this problem, we propose a novel deep unsupervised hashing method, namely Distilled Smooth Guidance (DSG), which can learn a distilled dataset consisting of similarity signals as well as smooth confidence signals. To be specific, we obtain the similarity confidence weights based on the initial noisy similarity signals learned from local structures and construct a priority loss function for smooth similarity-preserving learning. Besides, global information based on clustering is utilized to distill the image pairs by removing contradictory similarity signals. Extensive experiments on three widely used benchmark datasets show that the proposed DSG consistently outperforms the state-of-the-art search methods.

cs.CV↗
arXiv2021

ASPER: Attention-based Approach to Extract Syntactic Patterns denoting Semantic Relations in Sentential Context

Semantic relationships, such as hyponym-hypernym, cause-effect, meronym-holonym etc. between a pair of entities in a sentence are usually reflected through syntactic patterns. Automatic extraction of such patterns benefits several downstream tasks, including, entity extraction, ontology building, and question answering. Unfortunately, automatic extraction of such patterns has not yet received much attention from NLP and information retrieval researchers. In this work, we propose an attention-based supervised deep learning model, ASPER, which extracts syntactic patterns between entities exhibiting a given semantic relation in the sentential context. We validate the performance of ASPER on three distinct semantic relations -- hyponym-hypernym, cause-effect, and meronym-holonym on six datasets. Experimental results show that for all these semantic relations, ASPER can automatically identify a collection of syntactic patterns reflecting the existence of such a relation between a pair of entities in a sentence. In comparison to the existing methodologies of syntactic pattern extraction, ASPER's performance is substantially superior.

cs.CL↗
arXiv2021

Normalized solutions to a class of Kirchhoff equations with Sobolev critical exponent

In this paper, we consider the existence and asymptotic properties of solutions to the following Kirchhoff equation \begin{equation}\label{1}\nonumber - \Bigl(a+b\int_{\R^3} {{{\left| {\nabla u} \right|}^2}}\Bigl) Δu =λu+ {| u |^{p - 2}}u+μ{| u |^{q - 2}}u \text { in } \mathbb{R}^{3} \end{equation} under the normalized constraint $\int_{{\mathbb{R}^3}} {{u}^2}=c^2$, where $a\!>\!0$, $b\!>\!0$, $c\!>\!0$, $2\!<\!q\!<\!\frac{14}{3}\!<\! p\!\leq\!6$ or $\frac{14}{3}\!<\!q\!<\! p\!\leq\! 6$, $μ\!>\!0$ and $λ\!\in\!\R$ appears as a Lagrange multiplier. In both cases for the range of $p$ and $q$, the Sobolev critical exponent $p\!=\!6$ is involved and the corresponding energy functional is unbounded from below on $S_c=\Big\{ u \in H^{1}({\mathbb{R}^3}): \int_{{\mathbb{R}^3}} {{u}^2}=c^2 \Big\}$. If $2\!<\!q\!<\!\frac{10}{3}$ and $\frac{14}{3}\!<\! p\!<\!6$, we obtain a multiplicity result to the equation. If $2\!<\!q\!<\!\frac{10}{3}\!<\! p\!=\!6$ or $\frac{14}{3}\!<\!q\!<\! p\!\leq\! 6$, we get a ground state solution to the equation. Furthermore, we derive several asymptotic results on the obtained normalized solutions. Our results extend the results of N. Soave (J. Differential Equations 2020 $\&$ J. Funct. Anal. 2020), which studied the nonlinear Schrödinger equations with combined nonlinearities, to the Kirchhoff equations. To deal with the special difficulties created by the nonlocal term $({\int_{\R^3} {\left| {\nabla u} \right|} ^2}) Δu$ appearing in Kirchhoff type equations, we develop a perturbed Pohozaev constraint approach and we find a way to get a clear picture of the profile of the fiber map via careful analysis. In the meantime, we need some subtle energy estimates under the $L^2$-constraint to recover compactness in the Sobolev critical case.

math.AP↗
arXiv2021

Multiplicity and asymptotics of standing waves for the energy critical half-wave

In this paper, we consider the multiplicity and asymptotics of standing waves with prescribed mass $\int_{{\mathbb{R}^N}} {{u}^2}=a^2$ to the energy critical half-wave \begin{equation}\label{eqA0.1} \sqrt{-Δ}u=λu+μ|u|^{q-2} u+|u|^{2^*-2}u,\ \ u\in H^{1/2}(\R^N), \end{equation} where $N\!\geq\! 2$, $a\!>\!0$, $q \!\in\!\big(2,2+\frac{2}{N}\big)$, $2^*\!=\!\frac{2N}{N-1}$ and $λ\!\in\!\R$ appears as a Lagrange multiplier. We show that \eqref{eqA0.1} admits a ground state $u_a$ and an excited state $v_a$, which are characterised by a local minimizer and a mountain-pass critical point of the corresponding energy functional. Several asymptotic properties of $\{u_a\}$, $\{v_a\}$ are obtained and it is worth pointing out that we get a precise description of $\{u_a\}$ as $a\!\to\! 0^+$ without needing any uniqueness condition on the related limit problem. The main contribution of this paper is to extend the main results in J. Bellazzini et al. [Math. Ann. 371 (2018), 707-740] from energy subcritical to energy critical case. Furthermore, these results can be extended to the general fractional nonlinear Schrödinger equation with Sobolev critical exponent, which generalize the work of H. J. Luo-Z. T. Zhang [Calc. Var. Partial Differ. Equ. 59 (2020)] from energy subcritical to energy critical case.

math.AP↗
arXiv2021

The Energy Efficiency of Interfacial Solar Desalination: Insights from Detailed Theoretical Analysis

Solar-thermal evaporation, a traditional steam generation method for solar desalination, has received numerous attentions in recent years due to the significant increase in efficiency by adopting interfacial evaporation. While most of the previous studies focus on improving the evaporation efficiency by materials innovation and system design, the underlying mechanisms of its energy efficiency are less explored, leading to many confusions and misunderstandings. Herein, we clarify these mechanisms with a detailed thermal analysis model. Using this model, we elucidate the advantages of interfacial evaporation over the traditional evaporation method. Furthermore, we clarify the role of tuning the solar flux and surface area on the evaporation efficiency. Moreover, we quantitatively prove that the influence of environmental conditions on evaporation efficiency could not be eliminated by subtracting the dark evaporation rate from evaporation rate under solar. We also find that interfacial evaporation in a solar still does not have the high overall solar desalination efficiency as expected, but further improvement is possible from the system design part. Our analysis gains insights to the thermal processes involved in interfacial solar evaporation and offers perspectives to the further development of interfacial solar desalination technology.

physics.geo-ph↗
arXiv2020

Normalized solutions for a fourth-order Schrödinger equation with positive second-order dispersion coefficient

We are concerned with the existence and asymptotic properties of solutions to the following fourth-order Schrödinger equation \begin{equation}\label{1} Δ^{2}u+μΔu-λu={|u|}^{p-2}u, ~~~~x \in \R^{N}\\ \end{equation} under the normalized constraint $$\int_{{\mathbb{R}^N}} {{u}^2}=a^2,$$ where $N\!\geq\!2$, $a,μ\!>\!0$, $2+\frac{8}{N}\!<\!p\!<\! 4^{*}\!=\!\frac{2N}{(N-4)^{+}}$ and $λ\in\R$ appears as a Lagrange multiplier. Since the second-order dispersion term affects the structure of the corresponding energy functional $$ E_μ(u)=\frac{1}{2}{||Δu||}_2^2-\fracμ{2}{||\nabla u||}_2^2-\frac{1}{p}{||u||}_p^p $$ we could find at least two normalized solutions to (\ref{1}) if $2\!+\!\frac{8}{N}\!<\! p\!<\!{ 4^{*} }$ and $μ^{pγ_p-2}a^{p-2}\!<\!C$ for some explicit constant $C\!=\!C(N,p)\!>\!0$ and $γ_p\!=\!\frac{N(p\!-\!2)}{4p}$. Furthermore, we give some asymptotic properties of the normalized solutions to (\ref{1}) as $μ\to0^+$ and $a\to0^+$, respectively. In conclusion, we mainly extend the results in \cite{DBon,dbJB}, which deal with (\ref{1}), from $μ\leq0$ to the case of $μ>0$, and also extend the results in \cite{TJLu,Nbal}, which deal with (\ref{1}), from $L^2$-subcritical and $L^2$-critical setting to $L^2$-supercritical setting.

math.AP↗
arXiv2020

Ground states for 3D dipolar Bose-Einstein condensate involving quantum fluctuations and three-body losses

We consider ground states of three-dimensional dipolar Bose-Einstein condensate involving quantum fluctuations and three-body losses, which can be described equivalently by positive $L^2$-constraint critical point of the Gross-Pitaevskii energy functional \[E(u)\!=\!\frac{1}{2}\int_{{\mathbb{R}^3}} {|\nabla u|}^2dx+\frac{λ_{1}}{2}\int_{{\mathbb{R}^3}} {| u|}^4dx+\frac{λ_{2}}{2} \int_{\mathbb{R}^{3}}\left(K \star|u|^{2}\right)|u|^{2} d x+\frac{2λ_{3}}{p}\int_{{\mathbb{R}^3}} {|u|}^{p}dx,\] where $2 \!0$ with $V^c_{R_0} \!:=\!\left\{u \!\in\! S_c : \big(\int_{{\mathbb{R}^3}} {{|\nabla u|}^2dx}\big)^{\frac{1}{2}} \!<\!R_0\right\}$. We show that $m(c,R_0)$ is achieved by some $u_c>0$, which is a stable ground state. Furthermore, by refining the upper bound of $m(c, R_0)$, we provide a precise description of the asymptotic behavior of $u_c$ as the mass $c$ vanishes, i.e. $${[\frac{{p|{λ_3}|}}{2{γ_c}}]^{\frac{1}{p - 2}}}{u_c}(\frac{x + {y_c}}{{\sqrt {2{δ_p}{γ_c}} }}) \to {W_p}\;\;\;\;{\rm{in}}\;\;\;\;{H^1}({\mathbb{R}^3})\;\;\;\;{\rm{for some}}\;\;\;\;{y_c} \in {\mathbb{R}^3}\;\;\;\;{\rm{as}}\;\;\;\;c \to {0^ + }.$$

math.AP↗
arXiv2020

Multiplicity, asymptotics and stability of standing waves for nonlinear Schrödinger equation with rotation

In this article, we study the multiplicity, asymptotics and stability of standing waves with prescribed mass $c>0$ for nonlinear Schrödinger equation with rotation in the mass-supercritical regime arising in Bose-Einstein condensation. Under suitable restriction on the rotation frequency, by searching critical points of the corresponding energy functional on the mass-sphere, we obtain a local minimizer $u_c$ and a mountain pass solution $\hat{u}_c$. %under suitable assumptions on the related parameters. Furthermore, we show that $u_c$ is a ground state for small mass $c>0$ and describe a mass collapse behavior of the minimizers as $c\to 0$, while $\hat{u}_c$ is an excited state. Finally, we prove that the standing wave associated with $u_c$ is stable. Notice that the pioneering works \cite{aMsC,shYZ} imply that finite time blow-up of solutions to this model occurred in the mass-supercritical setting, therefore, we in the present paper obtain a new stability result. The main contribution of this paper is to extend the main results in \cite{JeSp,gYlW} concerning the same model from mass-subcritical and mass-critical regimes to mass-supercritical regime, where the physically most relevant case is covered.

math.AP↗