arXiv subjects

Xiaoguang Li

Publications and source records attributed to Xiaoguang Li.

At least 91 records · Page 5Linked to original sources

arXiv2021

JPGNet: Joint Predictive Filtering and Generative Network for Image Inpainting

Image inpainting aims to restore the missing regions of corrupted images and make the recovery result identical to the originally complete image, which is different from the common generative task emphasizing the naturalness or realism of generated images. Nevertheless, existing works usually regard it as a pure generation problem and employ cutting-edge deep generative techniques to address it. The generative networks can fill the main missing parts with realistic contents but usually distort the local structures or introduce obvious artifacts. In this paper, for the first time, we formulate image inpainting as a mix of two problems, predictive filtering and deep generation. Predictive filtering is good at preserving local structures and removing artifacts but falls short to complete the large missing regions. The deep generative network can fill the numerous missing pixels based on the understanding of the whole scene but hardly restores the details identical to the original ones. To make use of their respective advantages, we propose the joint predictive filtering and generative network (JPGNet) that contains three branches: predictive filtering & uncertainty network (PFUNet), deep generative network, and uncertainty-aware fusion network (UAFNet). The PFUNet can adaptively predict pixel-wise kernels for filtering-based inpainting according to the input image and output an uncertainty map. This map indicates the pixels should be processed by filtering or generative networks, which is further fed to the UAFNet for a smart combination between filtering and generative results. Note that, our method as a novel inpainting framework can benefit any existing generation-based methods. We validate our method on three public datasets, Dunhuang, Places2, and CelebA, and demonstrate that our method can enhance three state-of-the-art generative methods significantly with slightly extra time costs.

cs.CV↗

arXiv2021

Unsupervised Open-Domain Question Answering

Open-domain Question Answering (ODQA) has achieved significant results in terms of supervised learning manner. However, data annotation cannot also be irresistible for its huge demand in an open domain. Though unsupervised QA or unsupervised Machine Reading Comprehension (MRC) has been tried more or less, unsupervised ODQA has not been touched according to our best knowledge. This paper thus pioneers the work of unsupervised ODQA by formally introducing the task and proposing a series of key data construction methods. Our exploration in this work inspiringly shows unsupervised ODQA can reach up to 86% performance of supervised ones.

cs.CL↗

arXiv2021

Transition path theory for Langevin dynamics on manifold: optimal control and data-driven solver

We present a data-driven point of view for rare events, which represent conformational transitions in biochemical reactions modeled by over-damped Langevin dynamics on manifolds in high dimensions. We first reinterpret the transition state theory and the transition path theory from the optimal control viewpoint. Given point clouds sampled from a reaction dynamics, we construct a discrete Markov process based on an approximated Voronoi tesselation. We use the constructed Markov process to compute a discrete committor function whose level set automatically orders the point clouds. Then based on the committor function, an optimally controlled random walk on point clouds is constructed and utilized to efficiently sample transition paths, which become an almost sure event in $O(1)$ time instead of a rare event in the original reaction dynamics. To compute the mean transition path efficiently, a local averaging algorithm based on the optimally controlled random walk is developed, which adapts the finite temperature string method to the controlled Monte Carlo samples. Numerical examples on sphere/torus including a conformational transition for the alanine dipeptide in vacuum are conducted to illustrate the data-driven solver for the transition path theory on point clouds. The mean transition path obtained via the controlled Monte Carlo simulations highly coincides with the computed dominant transition path in the transition path theory.

math.OC↗

arXiv2021

Non-invasive Self-attention for Side Information Fusion in Sequential Recommendation

Sequential recommender systems aim to model users' evolving interests from their historical behaviors, and hence make customized time-relevant recommendations. Compared with traditional models, deep learning approaches such as CNN and RNN have achieved remarkable advancements in recommendation tasks. Recently, the BERT framework also emerges as a promising method, benefited from its self-attention mechanism in processing sequential data. However, one limitation of the original BERT framework is that it only considers one input source of the natural language tokens. It is still an open question to leverage various types of information under the BERT framework. Nonetheless, it is intuitively appealing to utilize other side information, such as item category or tag, for more comprehensive depictions and better recommendations. In our pilot experiments, we found naive approaches, which directly fuse types of side information into the item embeddings, usually bring very little or even negative effects. Therefore, in this paper, we propose the NOninVasive self-attention mechanism (NOVA) to leverage side information effectively under the BERT framework. NOVA makes use of side information to generate better attention distribution, rather than directly altering the item embedding, which may cause information overwhelming. We validate the NOVA-BERT model on both public and commercial datasets, and our method can stably outperform the state-of-the-art models with negligible computational overheads.

cs.IR↗

arXiv2020

HopRetriever: Retrieve Hops over Wikipedia to Answer Complex Questions

Collecting supporting evidence from large corpora of text (e.g., Wikipedia) is of great challenge for open-domain Question Answering (QA). Especially, for multi-hop open-domain QA, scattered evidence pieces are required to be gathered together to support the answer extraction. In this paper, we propose a new retrieval target, hop, to collect the hidden reasoning evidence from Wikipedia for complex question answering. Specifically, the hop in this paper is defined as the combination of a hyperlink and the corresponding outbound link document. The hyperlink is encoded as the mention embedding which models the structured knowledge of how the outbound link entity is mentioned in the textual context, and the corresponding outbound link document is encoded as the document embedding representing the unstructured knowledge within it. Accordingly, we build HopRetriever which retrieves hops over Wikipedia to answer complex questions. Experiments on the HotpotQA dataset demonstrate that HopRetriever outperforms previously published evidence retrieval methods by large margins. Moreover, our approach also yields quantifiable interpretations of the evidence collection process.

cs.CL↗

arXiv2020

Monitoring-based Differential Privacy Mechanism Against Query-Flooding Parameter Duplication Attack

Public intelligent services enabled by machine learning algorithms are vulnerable to model extraction attacks that can steal confidential information of the learning models through public queries. Though there are some protection options such as differential privacy (DP) and monitoring, which are considered promising techniques to mitigate this attack, we still find that the vulnerability persists. In this paper, we propose an adaptive query-flooding parameter duplication (QPD) attack. The adversary can infer the model information with black-box access and no prior knowledge of any model parameters or training data via QPD. We also develop a defense strategy using DP called monitoring-based DP (MDP) against this new attack. In MDP, we first propose a novel real-time model extraction status assessment scheme called Monitor to evaluate the situation of the model. Then, we design a method to guide the differential privacy budget allocation called APBA adaptively. Finally, all DP-based defenses with MDP could dynamically adjust the amount of noise added in the model response according to the result from Monitor and effectively defends the QPD attack. Furthermore, we thoroughly evaluate and compare the QPD attack and MDP defense performance on real-world models with DP and monitoring protection.

cs.CR↗

arXiv2020

SparTerm: Learning Term-based Sparse Representation for Fast Text Retrieval

Term-based sparse representations dominate the first-stage text retrieval in industrial applications, due to its advantage in efficiency, interpretability, and exact term matching. In this paper, we study the problem of transferring the deep knowledge of the pre-trained language model (PLM) to Term-based Sparse representations, aiming to improve the representation capacity of bag-of-words(BoW) method for semantic-level matching, while still keeping its advantages. Specifically, we propose a novel framework SparTerm to directly learn sparse text representations in the full vocabulary space. The proposed SparTerm comprises an importance predictor to predict the importance for each term in the vocabulary, and a gating controller to control the term activation. These two modules cooperatively ensure the sparsity and flexibility of the final text representation, which unifies the term-weighting and expansion in the same framework. Evaluated on MSMARCO dataset, SparTerm significantly outperforms traditional sparse methods and achieves state of the art ranking performance among all the PLM-based sparse models.

cs.IR↗

arXiv2020

DUMA: Reading Comprehension with Transposition Thinking

Multi-choice Machine Reading Comprehension (MRC) requires model to decide the correct answer from a set of answer options when given a passage and a question. Thus in addition to a powerful Pre-trained Language Model (PrLM) as encoder, multi-choice MRC especially relies on a matching network design which is supposed to effectively capture the relationships among the triplet of passage, question and answers. While the newer and more powerful PrLMs have shown their mightiness even without the support from a matching network, we propose a new DUal Multi-head Co-Attention (DUMA) model, which is inspired by human's transposition thinking process solving the multi-choice MRC problem: respectively considering each other's focus from the standpoint of passage and question. The proposed DUMA has been shown effective and is capable of generally promoting PrLMs. Our proposed method is evaluated on two benchmark multi-choice MRC tasks, DREAM and RACE, showing that in terms of powerful PrLMs, DUMA can still boost the model to reach new state-of-the-art performance.

cs.CL↗

arXiv2020

Blur-Attention: A boosting mechanism for non-uniform blurred image restoration

Dynamic scene deblurring is a challenging problem in computer vision. It is difficult to accurately estimate the spatially varying blur kernel by traditional methods. Data-driven-based methods usually employ kernel-free end-to-end mapping schemes, which are apt to overlook the kernel estimation. To address this issue, we propose a blur-attention module to dynamically capture the spatially varying features of non-uniform blurred images. The module consists of a DenseBlock unit and a spatial attention unit with multi-pooling feature fusion, which can effectively extract complex spatially varying blur features. We design a multi-level residual connection structure to connect multiple blur-attention modules to form a blur-attention network. By introducing the blur-attention network into a conditional generation adversarial framework, we propose an end-to-end blind motion deblurring method, namely Blur-Attention-GAN (BAG), for a single image. Our method can adaptively select the weights of the extracted features according to the spatially varying blur features, and dynamically restore the images. Experimental results show that the deblurring capability of our method achieved outstanding objective performance in terms of PSNR, SSIM, and subjective visual quality. Furthermore, by visualizing the features extracted by the blur-attention module, comprehensive discussions are provided on its effectiveness.

eess.IV↗

arXiv2020

Dynamics of ferromagnetic bimerons driven by spin currents and magnetic fields

Magnetic bimeron composed of two merons is a topological counterpart of magnetic skyrmion in in-plane magnets, which can be used as the nonvolatile information carrier in spintronic devices. Here we analytically and numerically study the dynamics of ferromagnetic bimerons driven by spin currents and magnetic fields. Numerical simulations demonstrate that two bimerons with opposite signs of topological numbers can be created simultaneously in a ferromagnetic thin film via current-induced spin torques. The current-induced spin torques can also drive the bimeron and its speed is analytically derived, which agrees with the numerical results. Since the bimerons with opposite topological numbers can coexist and have opposite drift directions, two-lane racetracks can be built in order to accurately encode the data bits. In addition, the dynamics of bimerons induced by magnetic field gradients and alternating magnetic fields are investigated. It is found that the bimeron driven by alternating magnetic fields can propagate along a certain direction. Moreover, combining a suitable magnetic field gradient, the Magnus-force-induced transverse motion can be completely suppressed, which implies that there is no skyrmion Hall effect. Our results are useful for understanding of the bimeron dynamics and may provide guidelines for building future bimeron-based spintronic devices.

cond-mat.mes-hall↗

arXiv2020

A lateral semicircular canal segmentation based geometric calibration for human temporal bone CT Image

Computed Tomography (CT) of the temporal bone has become an important method for diagnosing ear diseases. Due to the different posture of the subject and the settings of CT scanners, the CT image of the human temporal bone should be geometrically calibrated to ensure the symmetry of the bilateral anatomical structure. Manual calibration is a time-consuming task for radiologists and an important pre-processing step for further computer-aided CT analysis. We propose an automatic calibration algorithm for temporal bone CT images. The lateral semicircular canals (LSCs) are segmented as anchors at first. Then, we define a standard 3D coordinate system. The key step is the LSC segmentation. We design a novel 3D LSC segmentation encoder-decoder network, which introduces a 3D dilated convolution and a multi-pooling scheme for feature fusion in the encoding stage. The experimental results show that our LSC segmentation network achieved a higher segmentation accuracy. Our proposed method can help to perform calibration of temporal bone CT images efficiently.

eess.IV↗

arXiv2020

Mitigating Query-Flooding Parameter Duplication Attack on Regression Models with High-Dimensional Gaussian Mechanism

Public intelligent services enabled by machine learning algorithms are vulnerable to model extraction attacks that can steal confidential information of the learning models through public queries. Differential privacy (DP) has been considered a promising technique to mitigate this attack. However, we find that the vulnerability persists when regression models are being protected by current DP solutions. We show that the adversary can launch a query-flooding parameter duplication (QPD) attack to infer the model information by repeated queries. To defend against the QPD attack on logistic and linear regression models, we propose a novel High-Dimensional Gaussian (HDG) mechanism to prevent unauthorized information disclosure without interrupting the intended services. In contrast to prior work, the proposed HDG mechanism will dynamically generate the privacy budget and random noise for different queries and their results to enhance the obfuscation. Besides, for the first time, HDG enables an optimal privacy budget allocation that automatically determines the minimum amount of noise to be added per user-desired privacy level on each dimension. We comprehensively evaluate the performance of HDG using real-world datasets and shows that HDG effectively mitigates the QPD attack while satisfying the privacy requirements. We also prepare to open-source the relevant codes to the community for further research.

cs.CR↗

arXiv2020

The Graph Limit of The Minimizer of The Onsager-Machlup Functional and Its Computation

The Onsager-Machlup (OM) functional is well-known for characterizing the most probable transition path of a diffusion process with non-vanishing noise. However, it suffers from a notorious issue that the functional is unbounded below when the specified transition time $T$ goes to infinity. This hinders the interpretation of the results obtained by minimizing the OM functional. We provide a new perspective on this issue. Under mild conditions, we show that although the infimum of the OM functional becomes unbounded when $T$ goes to infinity, the sequence of minimizers does contain convergent subsequences on the space of curves. The graph limit of this minimizing subsequence is an extremal of the abbreviated action functional, which is related to the OM functional via the Maupertuis principle with an optimal energy. We further propose an energy-climbing geometric minimization algorithm (EGMA) which identifies the optimal energy and the graph limit of the transition path simultaneously. This algorithm is successfully applied to several typical examples in rare event studies. Some interesting comparisons with the Freidlin-Wentzell action functional are also made.

math.PR↗

arXiv2019

A Phase Shift Deep Neural Network for High Frequency Approximation and Wave Problems

In this paper, we propose a phase shift deep neural network (PhaseDNN), which provides a uniform wideband convergence in approximating high frequency functions and solutions of wave equations. The PhaseDNN makes use of the fact that common DNNs often achieve convergence in the low frequency range first, and a series of moderately-sized DNNs are constructed and trained for selected high frequency ranges. With the help of phase shifts in the frequency domain, each of the DNNs will be trained to approximate the function's higher frequency content over a specific range at the the speed of convergence as in the low frequency range. As a result, the proposed PhaseDNN is able to convert high frequency learning to low frequency one, allowing a uniform learning to wideband functions. The PhaseDNN will then be applied to find the solution of high frequency wave equations in inhomogeneous media through both differential and integral equation formulations with least square residual loss functions. Numerical results have demonstrated the capability of the PhaseDNN in learning high frequency functions and oscillatory solutions of interior and exterior Helmholtz equations.

cs.LG↗

arXiv2019

PhaseDNN - A Parallel Phase Shift Deep Neural Network for Adaptive Wideband Learning

In this paper, we propose a phase shift deep neural network (PhaseDNN) which provides a wideband convergence in approximating a high dimensional function during its training of the network. The PhaseDNN utilizes the fact that many DNN achieves convergence in the low frequency range first, thus, a series of moderately-sized of DNNs are constructed and trained in parallel for ranges of higher frequencies. With the help of phase shifts in the frequency domain, implemented through a simple phase factor multiplication on the training data, each DNN in the series will be trained to approximate the target function's higher frequency content over a specific range. Due to the phase shift, each DNN achieves the speed of convergence as in the low frequency range. As a result, the proposed PhaseDNN system is able to convert wideband frequency learning to low frequency learning, thus allowing a uniform learning to wideband high dimensional functions with frequency adaptive training. Numerical results have demonstrated the capability of PhaseDNN in learning information of a target function from low to high frequency uniformly.

eess.SP↗

arXiv2016

Enhanced Superconductivity in TiO Epitaxial Thin Films

Titanium oxides have many fascinating optical and electrical properties, such as the superconductivity at 2.0 K in cubic titanium monoxide TiO polycrystalline bulk. However, the lack of TiO single crystals or epitaxial films has prevented systematic investigations on its superconductivity. Here, we report the basic superconductivity characterizations of cubic TiO films epitaxially grown on (0001)-oriented Al2O3 substrates. The magnetic and electronic transport measurements confirmed that TiO is a type-II superconductor and the record high Tc is about 7.4 K. The lower critical field (Hc1) at 1.9 K, the extrapolated upper critical field Hc2(0) and coherence length are about 18 Oe, 13.7 T and 4.9 nm, respectively. With increasing pressure, the value of Tc shifts to lower temperature while the normal state resistivity increases. Our results on the superconducting TiO films confirm the strategy to achieve higher Tc in epitaxial films, which may be helpful for finding more superconducting materials in various related systems.

cond-mat.supr-con↗

arXiv2015

Induced effects by the substitution of Zn in Cu2ZnSnX4 (X = S and Se)

Based on the density functional theory with hybrid functional approach, we have studied the structural and thermodynamic stabilities of Cu2MSnX4 (M = Zn, Mg, and Ca; X = S and Se) alloy, and have further investigated the electronic and optical properties of stable Cu2MgSnS4 and Cu2MgSnSe4 phases. Thermal stability analysis indicates that Cu2MgSnS4 and Cu2MgSnSe4 are thermodynamically stable, while Cu2CaSnS4 and Cu2CaSnSe4 are unstable. The ground state configuration of the compound changes from kesterite into stannite structure when Zn atoms are substitued by larger Mg or Ca atoms. An energy separation between stannite and kesterite phase similar to that of CZTS is observed. Calculated electronic structures and optical properties suggest that Cu2MgSnS4 and Cu2MgSnSe4 can be efficient photovoltaic materials.

cond-mat.mtrl-sci↗

arXiv2015

Transition metals doped CuAlSe2 for promising intermediate band materials

Introducing an isolated intermediate band (IB) into a wide band gap semiconductor can potentially improve the optical absorption of the material beyond the Shockley-Queisser limitation for solar cells. Here, we present a systematic study of the thermodynamic stability, electronic structures, and optical properties of transition metals (M = Ti, V, and Fe) doped CuAlSe2 for potential IB thin film solar cells, by adopting the first-principles calculation based on the hybrid functional method. We found from chemical potential analysis that for all dopants considered, the stable doped phase only exits when the Al atom is substituted. More importantly, with this substitution, the IB feature is determined by $3d$ electronic nature of M^{3+} ion, and the electronic configuration of 3d^1 can drive a optimum IB that possesses half-filled character and suitable subbandgap from valence band or conduction band. We further show that Ti-doped CuAlSe2 is the more promising candidate for IB materials since the resulted IB in it is half filled and extra absorption peaks occurs in the optical spectrum accompanied with a largely enhanced light absorption intensity. The result offers a understanding for IB induced by transition metals into CuAlSe2 and is significant to fabricate the related IB materials.

cond-mat.mtrl-sci↗