arXiv subjects

Jian Wu

Publications and source records attributed to Jian Wu.

At least 361 records · Page 20Linked to original sources

arXiv2019

Searching for the possible signal of the photon-axionlike particle oscillation in the combined GeV and TeV spectra of supernova remnants

The conversion between photons and axionlike particles (ALPs) in the Milky Way magnetic field could result in the detectable oscillation phenomena in $γ$-ray spectra of Galactic sources. In this work, the GeV (Fermi-LAT) and TeV (MAGIC/VERITAS/H.E.S.S.) data of three bright supernova remnants (SNRs, ie. IC443, W51C and W49B) have been adopted together to search such the oscillation effect. Different from our previous analysis of the sole Fermi-LAT data of IC443, we do not find any reliable signal for the photon-ALP oscillation in the joint broadband spectrum of each SNR. The reason for the inconsistence is that in this work we use the latest revision (P8R3) of Fermi-LAT data, updated diffuse emission templates and the new version of the source catalog (4FGL), which lead to some modification of the GeV spectrum of IC443. Then we set constraints on ALP parameters based on the combined analysis of all the three sources. Though these constraints are somewhat weaker than limits from the CAST experiment and globular clusters, they are supportive of and complementary to these other results.

astro-ph.HE↗

arXiv2019

Multi-Modal Attention Network Learning for Semantic Source Code Retrieval

Code retrieval techniques and tools have been playing a key role in facilitating software developers to retrieve existing code fragments from available open-source repositories given a user query. Despite the existing efforts in improving the effectiveness of code retrieval, there are still two main issues hindering them from being used to accurately retrieve satisfiable code fragments from large-scale repositories when answering complicated queries. First, the existing approaches only consider shallow features of source code such as method names and code tokens, but ignoring structured features such as abstract syntax trees (ASTs) and control-flow graphs (CFGs) of source code, which contains rich and well-defined semantics of source code. Second, although the deep learning-based approach performs well on the representation of source code, it lacks the explainability, making it hard to interpret the retrieval results and almost impossible to understand which features of source code contribute more to the final results. To tackle the two aforementioned issues, this paper proposes MMAN, a novel Multi-Modal Attention Network for semantic source code retrieval. A comprehensive multi-modal representation is developed for representing unstructured and structured features of source code, with one LSTM for the sequential tokens of code, a Tree-LSTM for the AST of code and a GGNN (Gated Graph Neural Network) for the CFG of code. Furthermore, a multi-modal attention fusion layer is applied to assign weights to different parts of each modality of source code and then integrate them into a single hybrid representation. Comprehensive experiments and analysis on a large-scale real-world dataset show that our proposed model can accurately retrieve code snippets and outperforms the state-of-the-art methods.

cs.SE↗

arXiv2019

Time Domain Audio Visual Speech Separation

Audio-visual multi-modal modeling has been demonstrated to be effective in many speech related tasks, such as speech recognition and speech enhancement. This paper introduces a new time-domain audio-visual architecture for target speaker extraction from monaural mixtures. The architecture generalizes the previous TasNet (time-domain speech separation network) to enable multi-modal learning and at meanwhile it extends the classical audio-visual speech separation from frequency-domain to time-domain. The main components of proposed architecture include an audio encoder, a video encoder that extracts lip embedding from video streams, a multi-modal separation network and an audio decoder. Experiments on simulated mixtures based on recently released LRS2 dataset show that our method can bring 3dB+ and 4dB+ Si-SNR improvements on two- and three-speaker cases respectively, compared to audio-only TasNet and frequency-domain audio-visual networks

eess.AS↗

arXiv2019

BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer

Modeling users' dynamic and evolving preferences from their historical behaviors is challenging and crucial for recommendation systems. Previous methods employ sequential neural networks (e.g., Recurrent Neural Network) to encode users' historical interactions from left to right into hidden representations for making recommendations. Although these methods achieve satisfactory results, they often assume a rigidly ordered sequence which is not always practical. We argue that such left-to-right unidirectional architectures restrict the power of the historical sequence representations. For this purpose, we introduce a Bidirectional Encoder Representations from Transformers for sequential Recommendation (BERT4Rec). However, jointly conditioning on both left and right context in deep bidirectional model would make the training become trivial since each item can indirectly "see the target item". To address this problem, we train the bidirectional model using the Cloze task, predicting the masked items in the sequence by jointly conditioning on their left and right context. Comparing with predicting the next item at each position in a sequence, the Cloze task can produce more samples to train a more powerful bidirectional model. Extensive experiments on four benchmark datasets show that our model outperforms various state-of-the-art sequential models consistently.

cs.IR↗

arXiv2019

An Optimized Quantum Maximum or Minimum Searching Algorithm and its Circuits

Finding a maximum or minimum is a fundamental building block in many mathematical models. Compared with classical algorithms, Durr, Hoyer's quantum algorithm (DHA) achieves quadratic speed. However, its key step, the quantum exponential searching algorithm (QESA), which is based on Grover algorithm, is not a sure-success algorithm. Meanwhile, quantum circuits encounter the gate decomposition problem due to variation of the scale of data. In this paper, we propose an optimized quantum algorithm for searching maximum and minimum, based on DHA and the optimal quantum exact search algorithm. Furthermore, we provide the corresponding quantum circuits, together with three equivalent simplifications. In circumstances when we can exactly estimate the ratio of the number of solutions M and the searched space N, our method can improve the successful probability close to 100%. Furthermore, compared with DHA, our algorithm shows an advantage in complexity with large databases and in the gate complexity of constructing oracles. Experiments have been executed on an IBM superconducting processor with two qubits, and a practical problem of finding the minimum from Titanic passengers' age was numerically simulated. Both showed that our optimized maximum or minimum performs more efficiently compared with DHA. Our algorithm can serve as an important subroutine in various quantum algorithms which involves searching maximum or minimum.

quant-ph↗

arXiv2019

Echo in a Single Molecule

Echo is a ubiquitous phenomenon found in many physical systems, ranging from spins in magnetic fields to particle beams in hadron accelerators. It is typically observed in inhomogeneously broadened ensembles of nonlinear objects, and is used to eliminate the effects of environmental-induced dephasing, enabling observation of proper, inherent object properties. Here, we report experimental observation of quantum wave packet echoes in a single isolated molecule. In contrast to conventional echoes, here the entire dephasing-rephasing cycle occurs within a single molecule without any inhomogeneous spread of molecular properties, or any interaction with the environment. In our experiments, we use a short laser pulse to impulsively excite a vibrational wave packet in an anharmonic molecular potential, and observe its oscillations and eventual dispersion with time. A second delayed pulsed excitation is applied, giving rise to an echo: a partial recovery of the initial coherent wavepacket. The vibrational dynamics of single molecules is visualized by time-delayed probe pulse dissociating them one at a time. Two mechanisms for the echo formation are discussed: ac Stark-induced molecular potential shaking and creation of depletion-induced "hole" in the nuclear spatial distribution. Interplay between the optically induced echoes and quantum revivals of the vibrational wave packets is observed and theoretically analyzed. The single molecule wave packet echoes may lead to the development of new tools for probing ultrafast intramolecular processes in various molecules.

physics.chem-ph↗

arXiv2019

Personalized Re-ranking for Recommendation

Ranking is a core task in recommender systems, which aims at providing an ordered list of items to users. Typically, a ranking function is learned from the labeled dataset to optimize the global performance, which produces a ranking score for each individual item. However, it may be sub-optimal because the scoring function applies to each item individually and does not explicitly consider the mutual influence between items, as well as the differences of users' preferences or intents. Therefore, we propose a personalized re-ranking model for recommender systems. The proposed re-ranking model can be easily deployed as a follow-up modular after any ranking algorithm, by directly using the existing ranking feature vectors. It directly optimizes the whole recommendation list by employing a transformer structure to efficiently encode the information of all items in the list. Specifically, the Transformer applies a self-attention mechanism that directly models the global relationships between any pair of items in the whole list. We confirm that the performance can be further improved by introducing pre-trained embedding to learn personalized encoding functions for different users. Experimental results on both offline benchmarks and real-world online e-commerce systems demonstrate the significant improvements of the proposed re-ranking model.

cs.IR↗

arXiv2019

A comprehensive study of speech separation: spectrogram vs waveform separation

Speech separation has been studied widely for single-channel close-talk microphone recordings over the past few years; developed solutions are mostly in frequency-domain. Recently, a raw audio waveform separation network (TasNet) is introduced for single-channel data, with achieving high Si-SNR (scale-invariant source-to-noise ratio) and SDR (source-to-distortion ratio) comparing against the state-of-the-art solution in frequency-domain. In this study, we incorporate effective components of the TasNet into a frequency-domain separation method. We compare both for alternative scenarios. We introduce a solution for directly optimizing the separation criterion in frequency-domain networks. In addition to speech separation objective and subjective measurements, we evaluate the separation performance on a speech recognition task as well. We study the speech separation problem for far-field data (more similar to naturalistic audio streams) and develop multi-channel solutions for both frequency and time-domain separators with utilizing spectral, spatial and speaker location information. For our experiments, we simulated multi-channel spatialized reverberate WSJ0-2mix dataset. Our experimental results show that spectrogram separation can achieve competitive performance with better network design. Multi-channel framework as well is shown to improve the single-channel performance relatively up to +35.5% and +46% in terms of WER and SDR, respectively.

cs.SD↗

arXiv2019

Dunhuang Grottoes Painting Dataset and Benchmark

This document introduces the background and the usage of the Dunhuang Grottoes Dataset and the benchmark. The documentation first starts with the background of the Dunhuang Grotto, which is widely recognised as an priceless heritage. Given that digital method is the modern trend for heritage protection and restoration. Follow the trend, we release the first public dataset for Dunhuang Grotto Painting restoration. The rest of the documentation details the painting data generation. To enable a data driven fashion, this dataset provided a large number of training and testing example which is sufficient for a deep learning approach. The detailed usage of the dataset as well as the benchmark is described.

cs.CV↗

arXiv2019

Cleaning Noisy and Heterogeneous Metadata for Record Linking Across Scholarly Big Datasets

Automatically extracted metadata from scholarly documents in PDF formats is usually noisy and heterogeneous, often containing incomplete fields and erroneous values. One common way of cleaning metadata is to use a bibliographic reference dataset. The challenge is to match records between corpora with high precision. The existing solution which is based on information retrieval and string similarity on titles works well only if the titles are cleaned. We introduce a system designed to match scholarly document entities with noisy metadata against a reference dataset. The blocking function uses the classic BM25 algorithm to find the matching candidates from the reference data that has been indexed by ElasticSearch. The core components use supervised methods which combine features extracted from all available metadata fields. The system also leverages available citation information to match entities. The combination of metadata and citation achieves high accuracy that significantly outperforms the baseline method on the same test dataset. We apply this system to match the database of CiteSeerX against Web of Science, PubMed, and DBLP. This method will be deployed in the CiteSeerX system to clean metadata and link records to other scholarly big datasets.

cs.DL↗

arXiv2019

End-to-End Multi-Channel Speech Separation

The end-to-end approach for single-channel speech separation has been studied recently and shown promising results. This paper extended the previous approach and proposed a new end-to-end model for multi-channel speech separation. The primary contributions of this work include 1) an integrated waveform-in waveform-out separation system in a single neural network architecture. 2) We reformulate the traditional short time Fourier transform (STFT) and inter-channel phase difference (IPD) as a function of time-domain convolution with a special kernel. 3) We further relaxed those fixed kernels to be learnable, so that the entire architecture becomes purely data-driven and can be trained from end-to-end. We demonstrate on the WSJ0 far-field speech separation task that, with the benefit of learnable spatial features, our proposed end-to-end multi-channel model significantly improved the performance of previous end-to-end single-channel method and traditional multi-channel methods.

cs.SD↗

arXiv2019

X2CT-GAN: Reconstructing CT from Biplanar X-Rays with Generative Adversarial Networks

Computed tomography (CT) can provide a 3D view of the patient's internal organs, facilitating disease diagnosis, but it incurs more radiation dose to a patient and a CT scanner is much more cost prohibitive than an X-ray machine too. Traditional CT reconstruction methods require hundreds of X-ray projections through a full rotational scan of the body, which cannot be performed on a typical X-ray machine. In this work, we propose to reconstruct CT from two orthogonal X-rays using the generative adversarial network (GAN) framework. A specially designed generator network is exploited to increase data dimension from 2D (X-rays) to 3D (CT), which is not addressed in previous research of GAN. A novel feature fusion method is proposed to combine information from two X-rays.The mean squared error (MSE) loss and adversarial loss are combined to train the generator, resulting in a high-quality CT volume both visually and quantitatively. Extensive experiments on a publicly available chest CT dataset demonstrate the effectiveness of the proposed method. It could be a nice enhancement of a low-cost X-ray machine to provide physicians a CT-like 3D volume in several niche applications.

eess.IV↗

arXiv2019

Improved Speaker-Dependent Separation for CHiME-5 Challenge

This paper summarizes several follow-up contributions for improving our submitted NWPU speaker-dependent system for CHiME-5 challenge, which aims to solve the problem of multi-channel, highly-overlapped conversational speech recognition in a dinner party scenario with reverberations and non-stationary noises. We adopt a speaker-aware training method by using i-vector as the target speaker information for multi-talker speech separation. With only one unified separation model for all speakers, we achieve a 10\% absolute improvement in terms of word error rate (WER) over the previous baseline of 80.28\% on the development set by leveraging our newly proposed data processing techniques and beamforming approach. With our improved back-end acoustic model, we further reduce WER to 60.15\% which surpasses the result of our submitted CHiME-5 challenge system without applying any fusion techniques.

eess.AS↗

arXiv2019

Practical Multi-fidelity Bayesian Optimization for Hyperparameter Tuning

Bayesian optimization is popular for optimizing time-consuming black-box objectives. Nonetheless, for hyperparameter tuning in deep neural networks, the time required to evaluate the validation error for even a few hyperparameter settings remains a bottleneck. Multi-fidelity optimization promises relief using cheaper proxies to such objectives --- for example, validation error for a network trained using a subset of the training points or fewer iterations than required for convergence. We propose a highly flexible and practical approach to multi-fidelity Bayesian optimization, focused on efficiently optimizing hyperparameters for iteratively trained supervised learning models. We introduce a new acquisition function, the trace-aware knowledge-gradient, which efficiently leverages both multiple continuous fidelity controls and trace observations --- values of the objective at a sequence of fidelities, available when varying fidelity using training iterations. We provide a provably convergent method for optimizing our acquisition function and show it outperforms state-of-the-art alternatives for hyperparameter tuning of deep neural networks and large-scale kernel learning.

cs.LG↗

arXiv2019

Deep learning-based phase control method for coherent beam combining and its application in generating orbital angular momentum beams

We incorporate deep learning (DL) into coherent beam combining (CBC) systems for the first time, to the best of our knowledge. Using a well-trained convolutional neural network DL model, the phase error in CBC systems could be accurately estimated and preliminarily compensated. Then, the residual phase error is further compensated by stochastic parallel gradient descent (SPGD) algorithms. The two-stage phase control strategy combined with DL and SPGD algorithms is validated to be a feasible and promising technique to alleviate the long-standing problem that the phase control bandwidth decreases as the number of array elements expands. Further investigation denotes that the proposed phase control technique could be employed to generate orbital angular momentum (OAM) beams with different orders by distinguishing the OAM beams of conjugated phase distributions.

physics.optics↗

arXiv2019

Calibration and Status of the 3D Imaging Calorimeter of DAMPE for Cosmic Ray Physics on Orbit

The DArk Matter Particle Explorer (DAMPE) developed in China was designed to search for evidence of dark matter particles by observing primary cosmic rays and gamma rays in the energy range from 5 GeV to 10 TeV. Since its launch in December 2015, a large quantity of data has been recorded. With the data set acquired during more than a year of operation in space, a precise time-dependent calibration for the energy measured by the BGO ECAL has been developed. In this report, the instrumentation and development of the BGO Electromagnetic Calorimeter (BGO ECAL) are briefly described. The calibration on orbit, including that of the pedestal, attenuation length, minimum ionizing particle peak, and dynode ratio, is discussed, and additional details about the calibration methods and performance in space are presented.

astro-ph.IM↗

arXiv2018

A protonated brownmillerite electrolyte for superior low-temperature proton conductivity

Design novel solid oxide electrolyte with enhanced ionic conductivity forms one of the Holy Grails in the field of materials science due to its great potential for wide range of energy applications. Conventional solid oxide electrolyte typically requires elevated temperature to activate the ionic transportation, while it has been increasing research interests to reduce the operating temperature due to the associated scientific and technological importance. Here, we report a conceptually new solid oxide electrolyte, HSrCoO2.5, which shows an exceptional enhanced proton conductivity at low temperature region (from room temperature to 140 oC). Combining both the experimental results and corresponding first-principles calculations, we attribute these intriguing properties to the extremely-high proton concentration as well as the well-ordered oxygen vacancy channels inherited from the novel crystalline structure of HSrCoO2.5. This result provides a new strategy to design novel solid oxide electrolyte with excellent proton conductivity for wide ranges of energy-related applications.

cond-mat.mtrl-sci↗

arXiv2018

Improving Automatic Source Code Summarization via Deep Reinforcement Learning

Code summarization provides a high level natural language description of the function performed by code, as it can benefit the software maintenance, code categorization and retrieval. To the best of our knowledge, most state-of-the-art approaches follow an encoder-decoder framework which encodes the code into a hidden space and then decode it into natural language space, suffering from two major drawbacks: a) Their encoders only consider the sequential content of code, ignoring the tree structure which is also critical for the task of code summarization, b) Their decoders are typically trained to predict the next word by maximizing the likelihood of next ground-truth word with previous ground-truth word given. However, it is expected to generate the entire sequence from scratch at test time. This discrepancy can cause an \textit{exposure bias} issue, making the learnt decoder suboptimal. In this paper, we incorporate an abstract syntax tree structure as well as sequential content of code snippets into a deep reinforcement learning framework (i.e., actor-critic network). The actor network provides the confidence of predicting the next word according to current state. On the other hand, the critic network evaluates the reward value of all possible extensions of the current state and can provide global guidance for explorations. We employ an advantage reward composed of BLEU metric to train both networks. Comprehensive experiments on a real-world dataset show the effectiveness of our proposed model when compared with some state-of-the-art methods.

cs.SE↗