arXiv subjects

Hui Liu

Publications and source records attributed to Hui Liu.

At least 487 records · Page 27Linked to original sources

arXiv2021

Extracting Concise Bug-Fixing Patches from Human-Written Patches in Version Control Systems

High-quality and large-scale repositories of real bugs and their concise patches collected from real-world applications are critical for research in software engineering community. In such a repository, each real bug is explicitly associated with its fix. Therefore, on one side, the real bugs and their fixes} may inspire novel approaches for finding, locating, and repairing software bugs; on the other side, the real bugs and their fixes are indispensable for rigorous and meaningful evaluation of approaches for software testing, fault localization, and program repair. To this end, a number of such repositories, e.g., Defects4J, have been proposed. However, such repositories are rather small because their construction involves expensive human intervention. Although bug-fixing code commits as well as associated test cases could be retrieved from version control systems automatically, existing approaches could not yet automatically extract concise bug-fixing patches from bug-fixing commits because such commits often involve bug-irrelevant changes. In this paper, we propose an automatic approach, called BugBuilder, to extracting complete and concise bug-fixing patches from human-written patches in version control systems. It excludes refactorings by detecting refactorings involved in bug-fixing commits, and reapplying detected refactorings on the faulty version. It enumerates all subsets of the remaining part and validates them on test cases. If none of the subsets has the potential to be a complete bug-fixing patch, the remaining part as a whole is taken as a complete and concise bug-fixing patch. Evaluation results on 809 real bug-fixing commits in Defects4J suggest that BugBuilder successfully generated complete and concise bug-fixing patches for forty percent of the bug-fixing commits, and its precision (99%) was even higher than human experts.

cs.SE↗

arXiv2021

Multiplicity and stability of closed characteristics on compact convex P-cyclic symmetric hypersurfaces in ${\bf R}^{2n}$

Let $Σ$ be a compact convex hypersurface in ${\bf R}^{2n}$ which is P-cyclic symmetric, i.e., $x\in Σ$ implies $Px\inΣ$ with P being a $2n\times2n$ symplectic orthogonal matrix and satisfying $P^k=I_{2n}$, $ker(P^l-I_{2n})=0$ for $1\leq l< k$, where $n, k\geq2$. In this paper, we prove that there exist at least $n$ geometrically distinct closed characteristics on $Σ$, which solves a longstanding conjecture about the multiplicity of closed characteristics for a broad class of compact convex hypersurfaces with symmetries(cf.,Page 235 of \cite{Eke1}). Based on the proof, we further prove that if the number of geometrically distinct closed characteristics on $Σ$ is finite, then at least $2[\frac{n}{2}]$ of them are non-hyperbolic; and if the number of geometrically distinct closed characteristics on $Σ$ is exactly $n$ and $k\geq3$, then all of them are P-cyclic symmetric, where a closed characteristic $(τ, y)$ on $Σ$ is called P-cyclic symmetric if $y({\bf R})=Py({\bf R})$.

math.DS↗

arXiv2021

RODNet: A Real-Time Radar Object Detection Network Cross-Supervised by Camera-Radar Fused Object 3D Localization

Various autonomous or assisted driving strategies have been facilitated through the accurate and reliable perception of the environment around a vehicle. Among the commonly used sensors, radar has usually been considered as a robust and cost-effective solution even in adverse driving scenarios, e.g., weak/strong lighting or bad weather. Instead of considering to fuse the unreliable information from all available sensors, perception from pure radar data becomes a valuable alternative that is worth exploring. In this paper, we propose a deep radar object detection network, named RODNet, which is cross-supervised by a camera-radar fused algorithm without laborious annotation efforts, to effectively detect objects from the radio frequency (RF) images in real-time. First, the raw signals captured by millimeter-wave radars are transformed to RF images in range-azimuth coordinates. Second, our proposed RODNet takes a sequence of RF images as the input to predict the likelihood of objects in the radar field of view (FoV). Two customized modules are also added to handle multi-chirp information and object relative motion. Instead of using human-labeled ground truth for training, the proposed RODNet is cross-supervised by a novel 3D localization of detected objects using a camera-radar fusion (CRF) strategy in the training stage. Finally, we propose a method to evaluate the object detection performance of the RODNet. Due to no existing public dataset available for our task, we create a new dataset, named CRUW, which contains synchronized RGB and RF image sequences in various driving scenarios. With intensive experiments, our proposed cross-supervised RODNet achieves 86% average precision and 88% average recall of object detection performance, which shows the robustness to noisy scenarios in various driving conditions.

cs.CV↗

arXiv2021

RODNet: Radar Object Detection Using Cross-Modal Supervision

Radar is usually more robust than the camera in severe driving scenarios, e.g., weak/strong lighting and bad weather. However, unlike RGB images captured by a camera, the semantic information from the radar signals is noticeably difficult to extract. In this paper, we propose a deep radar object detection network (RODNet), to effectively detect objects purely from the carefully processed radar frequency data in the format of range-azimuth frequency heatmaps (RAMaps). Three different 3D autoencoder based architectures are introduced to predict object confidence distribution from each snippet of the input RAMaps. The final detection results are then calculated using our post-processing method, called location-based non-maximum suppression (L-NMS). Instead of using burdensome human-labeled ground truth, we train the RODNet using the annotations generated automatically by a novel 3D localization method using a camera-radar fusion (CRF) strategy. To train and evaluate our method, we build a new dataset -- CRUW, containing synchronized videos and RAMaps in various driving scenarios. After intensive experiments, our RODNet shows favorable object detection performance without the presence of the camera.

cs.CV↗

arXiv2021

Twin-Field Quantum Key Distribution over 511 km Optical Fiber Linking two Distant Metropolitans

The basic principle of quantum mechanics guarantee the unconditional security of quantum key distribution (QKD) at the cost of inability of amplification of quantum state. As a result, despite remarkable progress in worldwide metropolitan QKD networks over the past decades, long haul fiber QKD network without trustful relay has not been achieved yet. Here, through sending-or-not-sending (SNS) protocol, we complete a twin field QKD (TF-QKD) and distribute secure keys without any trusted repeater over a 511 km long haul fiber trunk linking two distant metropolitans. Our secure key rate is around 3 orders of magnitudes greater than what is expected if the previous QKD field test system over the same length were applied. The efficient quantum-state transmission and stable single-photon interference over such a long distance deployed fiber paves the way to large-scale fiber quantum networks.

quant-ph↗

arXiv2021

Entanglement of two quantum memories via fibers over dozens of kilometres

Quantum internet will enable a number of revolutionary applications. It relies on entanglement of remote quantum memories over long distances. Despite enormous progresses so far, the maximal physical separation achieved between two nodes is 1.3 km, and challenges for long distance remain. Here we make a significant step forward by entangling two atomic ensembles in one lab via photon transmission through metropolitan-scale fibers. We use cavity enhancement to create bright atom-photon entanglement, and harness quantum frequency conversion to shift the atomic wavelength to telecom. We realize entanglement over 22 km field-deployed fibers via two-photon interference, and entanglement over 50 km coiled fibers via single-photon interference. Our experiment can be extended to physically separated nodes with similar distance as a functional segment for atomic quantum networks, thus paving the way towards establishing atomic entanglement over many nodes and over much longer distance.

quant-ph↗

arXiv2021

Field Test of Twin-Field Quantum Key Distribution through Sending-or-Not-Sending over 428 km

Quantum key distribution endows people with information-theoretical security in communications. Twin-field quantum key distribution (TF-QKD) has attracted considerable attention because of its outstanding key rates over long distances. Recently, several demonstrations of TF-QKD have been realized. Nevertheless, those experiments are implemented in the laboratory, remaining a critical question about whether the TF-QKD is feasible in real-world circumstances. Here, by adopting the sending-or-not-sending twin-field QKD (SNS-TF-QKD) with the method of actively odd parity pairing (AOPP), we demonstrate a field-test QKD over 428~km deployed commercial fiber and two users are physically separated by about 300~km in a straight line. To this end, we explicitly measure the relevant properties of the deployed fiber and develop a carefully designed system with high stability. The secure key rate we achieved breaks the absolute key rate limit of repeater-less QKD. The result provides a new distance record for the field test of both TF-QKD and all types of fiber-based QKD systems. Our work bridges the gap of QKD between laboratory demonstrations and practical applications, and paves the way for intercity QKD network with high-speed and measurement-device-independent security.

quant-ph↗

arXiv2020

Learning Light-Weight Translation Models from Deep Transformer

Recently, deep models have shown tremendous improvements in neural machine translation (NMT). However, systems of this kind are computationally expensive and memory intensive. In this paper, we take a natural step towards learning strong but light-weight NMT systems. We proposed a novel group-permutation based knowledge distillation approach to compressing the deep Transformer model into a shallow model. The experimental results on several benchmarks validate the effectiveness of our method. Our compressed model is 8X shallower than the deep model, with almost no loss in BLEU. To further enhance the teacher model, we present a Skipping Sub-Layer method to randomly omit sub-layers to introduce perturbation into training, which achieves a BLEU score of 30.63 on English-German newstest2014. The code is publicly available at https://github.com/libeineu/GPKD.

cs.CL↗

arXiv2020

Clustering Ensemble Meets Low-rank Tensor Approximation

This paper explores the problem of clustering ensemble, which aims to combine multiple base clusterings to produce better performance than that of the individual one. The existing clustering ensemble methods generally construct a co-association matrix, which indicates the pairwise similarity between samples, as the weighted linear combination of the connective matrices from different base clusterings, and the resulting co-association matrix is then adopted as the input of an off-the-shelf clustering algorithm, e.g., spectral clustering. However, the co-association matrix may be dominated by poor base clusterings, resulting in inferior performance. In this paper, we propose a novel low-rank tensor approximation-based method to solve the problem from a global perspective. Specifically, by inspecting whether two samples are clustered to an identical cluster under different base clusterings, we derive a coherent-link matrix, which contains limited but highly reliable relationships between samples. We then stack the coherent-link matrix and the co-association matrix to form a three-dimensional tensor, the low-rankness property of which is further explored to propagate the information of the coherent-link matrix to the co-association matrix, producing a refined co-association matrix. We formulate the proposed method as a convex constrained optimization problem and solve it efficiently. Experimental results over 7 benchmark data sets show that the proposed model achieves a breakthrough in clustering performance, compared with 12 state-of-the-art methods. To the best of our knowledge, this is the first work to explore the potential of low-rank tensor on clustering ensemble, which is fundamentally different from previous approaches.

cs.LG↗

arXiv2020

A nonlinear solar magnetic field calibration method for the filter-based magnetograph by the residual network

The method of solar magnetic field calibration for the filter-based magnetograph is normally the linear calibration method under weak-field approximation that cannot generate the strong magnetic field region well due to the magnetic saturation effect. We try to provide a new method to carry out the nonlinear magnetic calibration with the help of neural networks to obtain more accurate magnetic fields. We employed the data from Hinode/SP to construct a training, validation and test dataset. The narrow-band Stokes I, Q, U, and V maps at one wavelength point were selected from all the 112 wavelength points observed by SP so as to simulate the single-wavelength observations of the filter-based magnetograph. We used the residual network to model the nonlinear relationship between the Stokes maps and the vector magnetic fields. After an extensive performance analysis, it is found that the trained models could infer the longitudinal magnetic flux density, the transverse magnetic flux density, and the azimuth angle from the narrow-band Stokes maps with a precision comparable to the inversion results using 112 wavelength points. Moreover, the maps that were produced are much cleaner than the inversion results. The method can effectively overcome the magnetic saturation effect and infer the strong magnetic region much better than the linear calibration method. The residual errors of test samples to standard data are mostly about 50 G for both the longitudinal and transverse magnetic flux density. The values are about 100 G with our previous method of multilayer perceptron, indicating that the new method is more accurate in magnetic calibration.

astro-ph.SR↗

arXiv2020

Mid-InfraRed Outburst in Nearby Galaxies (MIRONG) I: Sample Selection and Characterization

The optical time-domain astronomy has grown rapidly in the past decade but the dynamic infrared sky is rarely explored. Aiming to construct a sample of mid-infrared outburst in nearby galaxies (MIRONG), we have conducted a systematical search of low-redshift ($z<0.35$) SDSS spectroscopic galaxies that have experienced recent MIR flares using their Wide-field Infrared Survey Explorer (WISE) light curves. A total of 137 galaxies have been selected by requiring a brightening amplitude of 0.5 magnitude in at least one WISE band with respect to their quiescent phases. Only a small faction (10.9%) has corresponding optical flares. Except for the four supernova (SNe) in our sample, the MIR luminosity of remaining sources ($L_{\rm 4.6μm}>10^{42}~\rm erg~s^{-1}$) are markedly brighter than known SNe and their physical locations are very close to the galactic center (median <0.1"). Only four galaxies are radio-loud indicating that synchrotron radiation from relativistic jets could contribute MIR variability. We propose that these MIR outburst are dominated by the dust echoes of transient accretion onto supermassive black holes, such as tidal disruption events (TDEs) and turn-on (changing-look) AGNs. Moreover, the inferred peak MIR luminosity function is generally consistent with the X-ray and optical TDEs at high end albeit with large uncertainties. Our results suggest that a large population of transients have been overlooked by optical surveys, probably due to dust obscuration or intrinsically optical weakness. Thus, a search in the infrared band is crucial for us to obtain a panoramic picture of nuclear outburst. The multiwavength follow-up observations of the MIRONG sample are in progress and will be presented in a series of subsequent papers.

astro-ph.GA↗

arXiv2020

Predicting the Evolution of Photospheric Magnetic Field in Solar Active Regions Using Deep Learning

The continuous observation of the magnetic field by Solar Dynamics Observatory (SDO)/Helioseismic and Magnetic Imager (HMI) produces numerous image sequences in time and space. These sequences provide data support for predicting the evolution of photospheric magnetic field. Based on the spatiotemporal long short-term memory(LSTM) network, we use the preprocessed data of photospheric magnetic field in active regions to build a prediction model for magnetic field evolution. Because of the elaborate learning and memory mechanism, the trained model can characterize the inherent relationships contained in spatiotemporal features. The testing results of the prediction model indicate that (1) the prediction pattern learned by the model can be applied to predict the evolution of new magnetic field in the next 6 hour that have not been trained, and predicted results are roughly consistent with real observed magnetic field evolution in terms of large-scale structure and movement speed; (2) the performance of the model is related to the prediction time; the shorter the prediction time, the higher the accuracy of the predicted results; (3) the performance of the model is stable not only for active regions in the north and south but also for data in positive and negative regions. Detailed experimental results and discussions on magnetic flux emergence and magnetic neutral lines finally show that the proposed model could effectively predict the large-scale and short-term evolution of the photospheric magnetic field in active regions. Moreover, our study may provide a reference for the spatiotemporal prediction of other solar activities.

astro-ph.SR↗

arXiv2020

Does Gender Matter? Towards Fairness in Dialogue Systems

Recently there are increasing concerns about the fairness of Artificial Intelligence (AI) in real-world applications such as computer vision and recommendations. For example, recognition algorithms in computer vision are unfair to black people such as poorly detecting their faces and inappropriately identifying them as "gorillas". As one crucial application of AI, dialogue systems have been extensively applied in our society. They are usually built with real human conversational data; thus they could inherit some fairness issues which are held in the real world. However, the fairness of dialogue systems has not been well investigated. In this paper, we perform a pioneering study about the fairness issues in dialogue systems. In particular, we construct a benchmark dataset and propose quantitative measures to understand fairness in dialogue models. Our studies demonstrate that popular dialogue models show significant prejudice towards different genders and races. Besides, to mitigate the bias in dialogue systems, we propose two simple but effective debiasing methods. Experiments show that our methods can reduce the bias in dialogue systems significantly. The dataset and the implementation are released to foster fairness research in dialogue systems.

cs.CL↗

arXiv2020

Mitigating Gender Bias for Neural Dialogue Generation with Adversarial Learning

Dialogue systems play an increasingly important role in various aspects of our daily life. It is evident from recent research that dialogue systems trained on human conversation data are biased. In particular, they can produce responses that reflect people's gender prejudice. Many debiasing methods have been developed for various NLP tasks, such as word embedding. However, they are not directly applicable to dialogue systems because they are likely to force dialogue models to generate similar responses for different genders. This greatly degrades the diversity of the generated responses and immensely hurts the performance of the dialogue models. In this paper, we propose a novel adversarial learning framework Debiased-Chat to train dialogue models free from gender bias while keeping their performance. Extensive experiments on two real-world conversation datasets show that our framework significantly reduces gender bias in dialogue models while maintaining the response quality. The implementation of the proposed framework is released.

cs.CL↗

arXiv2020

Geometry induced quantum Hall effect and Hall viscosity

For a particle confined to the two-dimensional helical surface embedded in four-dimensional (4D) Euclidean space, the effective Hamiltonian is deduced in the thin-layer quantization formalism. We find that the gauge structure of the effective dynamics is determined by torsion, which plays the role of U(1) gauge potential, and find that the topological structure of associated states is defined by orbital spin which originates from 4D space. Strikingly, the response to torsion contributes a quantum Hall effect, and the response to the deformation of torsion contributes Hall viscosity that is perfectly presented as a simultaneous occurrence of multiple channels for the quantum Hall effect. This result directly provides a way to probe Hall viscosity.

cond-mat.mes-hall↗

arXiv2020

Memory-efficient Embedding for Recommendations

Practical large-scale recommender systems usually contain thousands of feature fields from users, items, contextual information, and their interactions. Most of them empirically allocate a unified dimension to all feature fields, which is memory inefficient. Thus it is highly desired to assign different embedding dimensions to different feature fields according to their importance and predictability. Due to the large amounts of feature fields and the nuanced relationship between embedding dimensions with feature distributions and neural network architectures, manually allocating embedding dimensions in practical recommender systems can be very difficult. To this end, we propose an AutoML based framework (AutoDim) in this paper, which can automatically select dimensions for different feature fields in a data-driven fashion. Specifically, we first proposed an end-to-end differentiable framework that can calculate the weights over various dimensions for feature fields in a soft and continuous manner with an AutoML based optimization algorithm; then we derive a hard and discrete embedding component architecture according to the maximal weights and retrain the whole recommender framework. We conduct extensive experiments on benchmark datasets to validate the effectiveness of the AutoDim framework.

cs.IR↗

arXiv2020

Shallow-to-Deep Training for Neural Machine Translation

Deep encoders have been proven to be effective in improving neural machine translation (NMT) systems, but training an extremely deep encoder is time consuming. Moreover, why deep models help NMT is an open question. In this paper, we investigate the behavior of a well-tuned deep Transformer system. We find that stacking layers is helpful in improving the representation ability of NMT models and adjacent layers perform similarly. This inspires us to develop a shallow-to-deep training method that learns deep models by stacking shallow models. In this way, we successfully train a Transformer system with a 54-layer encoder. Experimental results on WMT'16 English-German and WMT'14 English-French translation tasks show that it is $1.4$ $\times$ faster than training from scratch, and achieves a BLEU score of $30.33$ and $43.29$ on two tasks. The code is publicly available at https://github.com/libeineu/SDT-Training/.

cs.CL↗

arXiv2020

Optimization of laser dynamics for active stabilization of DF--VECSELs dedicated to cesium CPT clocks

We report the implementation and performance of a double servo-loop for intensity and phase-difference active stabilization of a dual-frequency vertical external--cavity surface-emitting laser (DF-VECSEL) for coherent population trapping (CPT) of cesium atoms in the framework of compact atomic clocks. In--phase fully correlated pumping of the two laser modes is identified as the best scheme for intensity noise reduction, and an analytical model allows the optimization of the active stabilization strategy. Optical phase-locking the beat-note to a local oscillator leads to a phase noise level below -103~dBc/Hz at 100~Hz from the carrier. The laser contribution to the short-term frequency stability of the clock is predicted to be compatible with a targeted Allan deviation below $σ_y = 5\,\times 10^{-13}$ over one second.

physics.ins-det↗