arXiv subjects

Meng Liu

Publications and source records attributed to Meng Liu.

157 records · Page 9Linked to original sources

arXiv2020

Towards Deeper Graph Neural Networks

Graph neural networks have shown significant success in the field of graph representation learning. Graph convolutions perform neighborhood aggregation and represent one of the most important graph operations. Nevertheless, one layer of these neighborhood aggregation methods only consider immediate neighbors, and the performance decreases when going deeper to enable larger receptive fields. Several recent studies attribute this performance deterioration to the over-smoothing issue, which states that repeated propagation makes node representations of different classes indistinguishable. In this work, we study this observation systematically and develop new insights towards deeper graph neural networks. First, we provide a systematical analysis on this issue and argue that the key factor compromising the performance significantly is the entanglement of representation transformation and propagation in current graph convolution operations. After decoupling these two operations, deeper graph neural networks can be used to learn graph node representations from larger receptive fields. We further provide a theoretical analysis of the above observation when building very deep models, which can serve as a rigorous and gentle description of the over-smoothing issue. Based on our theoretical and empirical analysis, we propose Deep Adaptive Graph Neural Network (DAGNN) to adaptively incorporate information from large receptive fields. A set of experiments on citation, co-authorship, and co-purchase datasets have confirmed our analysis and insights and demonstrated the superiority of our proposed methods.

cs.LG↗

arXiv2020

Sparsely Grouped Multi-task Generative Adversarial Networks for Facial Attribute Manipulation

Recent Image-to-Image Translation algorithms have achieved significant progress in neural style transfer and image attribute manipulation tasks. However, existing approaches require exhaustively labelling training data, which is labor demanding, difficult to scale up, and hard to migrate into new domains. To overcome such a key limitation, we propose Sparsely Grouped Generative Adversarial Networks (SG-GAN) as a novel approach that can translate images on sparsely grouped datasets where only a few samples for training are labelled. Using a novel one-input multi-output architecture, SG-GAN is well-suited for tackling sparsely grouped learning and multi-task learning. The proposed model can translate images among multiple groups using only a single commonly trained model. To experimentally validate advantages of the new model, we apply the proposed method to tackle a series of attribute manipulation tasks for facial images. Experimental results demonstrate that SG-GAN can generate image translation results of comparable quality with baselines methods on adequately labelled datasets and results of superior quality on sparsely grouped datasets. The official implementation is publicly available:https://github.com/zhangqianhui/Sparsely-Grouped-GAN.

cs.CV↗

arXiv2020

NTIRE 2020 Challenge on Real Image Denoising: Dataset, Methods and Results

This paper reviews the NTIRE 2020 challenge on real image denoising with focus on the newly introduced dataset, the proposed methods and their results. The challenge is a new version of the previous NTIRE 2019 challenge on real image denoising that was based on the SIDD benchmark. This challenge is based on a newly collected validation and testing image datasets, and hence, named SIDD+. This challenge has two tracks for quantitatively evaluating image denoising performance in (1) the Bayer-pattern rawRGB and (2) the standard RGB (sRGB) color spaces. Each track ~250 registered participants. A total of 22 teams, proposing 24 methods, competed in the final phase of the challenge. The proposed methods by the participating teams represent the current state-of-the-art performance in image denoising targeting real noisy images. The newly collected SIDD+ datasets are publicly available at: https://bit.ly/siddplus_data.

cs.CV↗

arXiv2020

Hardware Impairments Aware Full-Duplex NOMA Networks Over Rician Fading Channels

A cooperative full duplex (FD) non-orthogonal multiple access (NOMA) scheme over Rician fading channels is considered. To be practical, imperfect successive interference cancellation (ipSIC) and residual hardware impairments (RHIs) at transceivers are taken into account. To evaluate the performance of the considered system, the analytical approximate expressions for the outage probability (OP) and the ergodic rate (ER) of the considered system are derived, and the asymptotic performance is also explored. Simulation results manifest that the FD cooperative NOMA can improve the ergodic sum rate (ESR) of the system compared to the half duplex (HD) mode.

cs.IT↗

arXiv2020

Microscopic aspects of artificial ageing in Al-Mg-Si alloys

Al-Mg-Si alloys with total solute contents ranging from 0.8 to 1.4 wt.% were solutionised, quenched and then artificially aged (AA) at 180 °C, after which positron annihilation lifetime spectroscopy was applied to obtain information about precipitation and vacancy evolution during preceding ageing. Hardness and electrical resistivity measurements were carried out to complement these measurements. AA was carried out in four different heating media, which allowed for varying the heating rate from 2.4 K/s to 170 K/s. The main result of the study is that there is a competition between vacancy losses and precipitation. Any precipitation taking place during quenching or during heating to the AA temperature helps to prevent vacancies from going to sinks and allows them to assist in solute clustering. Higher solute content, slower heating to 180 °C and natural pre-ageing before AA were found to have a comparable effect.

cond-mat.mtrl-sci↗

arXiv2019

KGAT: Knowledge Graph Attention Network for Recommendation

To provide more accurate, diverse, and explainable recommendation, it is compulsory to go beyond modeling user-item interactions and take side information into account. Traditional methods like factorization machine (FM) cast it as a supervised learning problem, which assumes each interaction as an independent instance with side information encoded. Due to the overlook of the relations among instances or items (e.g., the director of a movie is also an actor of another movie), these methods are insufficient to distill the collaborative signal from the collective behaviors of users. In this work, we investigate the utility of knowledge graph (KG), which breaks down the independent interaction assumption by linking items with their attributes. We argue that in such a hybrid structure of KG and user-item graph, high-order relations --- which connect two items with one or multiple linked attributes --- are an essential factor for successful recommendation. We propose a new method named Knowledge Graph Attention Network (KGAT) which explicitly models the high-order connectivities in KG in an end-to-end fashion. It recursively propagates the embeddings from a node's neighbors (which can be users, items, or attributes) to refine the node's embedding, and employs an attention mechanism to discriminate the importance of the neighbors. Our KGAT is conceptually advantageous to existing KG-based recommendation methods, which either exploit high-order relations by extracting paths or implicitly modeling them with regularization. Empirical results on three public benchmarks show that KGAT significantly outperforms state-of-the-art methods like Neural FM and RippleNet. Further studies verify the efficacy of embedding propagation for high-order relation modeling and the interpretability benefits brought by the attention mechanism.

cs.LG↗

arXiv2019

Bias Scheme Reducing Transient Currents and Speeding up Read Operations for 3-D Cross Point PCM

3-D cross point phase change memory (PCM) is a promising emerging memory. However, dynamic performances of 3-D cross point PCM are limited and the role of bias scheme is unknown. Previous studies on bias schemes for planar memories use static analyses to assess static array performances. Here, a high peak transient read current is found to result in a long read access time. Three factors which contribute to the high peak read current are analyzed. The proposed 2V/3 bias scheme and a single-reference parasitic-matching sensing circuit are utilized in a 64Mbit 3-D cross point PCM. The sensing time is reduced by 22.5%. Designing bias schemes through dynamic analyses paves the way for the development of high-performance 3-D PCM technology to boost the working efficiency of computing systems. The following contents are the listed errors as mentioned in the comments for reasons of withdrawal: missing labels of Fig.1, lack of the explanation of the peak current on the sensing speed, lack of the reason of the chosen of 2V/3 scheme, lack of influence of the parasitic capacitors, unclear explanation of the factor that contributes to a high peak read current, lack of comparisons, lack of the explanation of the sneak current, lack of the explanations of the SRPM sensing circuit, lack of the define valuables like NM_{A1}, unclear explanation of comparisons of the power consumption, wrong writing of the two conventional circuits, writing of 3D-IC, wrong writing of number of memory layers, etc. These errors may confuse a good understanding of our work. Therefore, we decide to withdraw this manuscript from arXiv. We will significantly rewrite it and publish the correct and complete paper in the future.

cs.ET↗

arXiv2019

Cost-Sensitive Feature Selection by Optimizing F-Measures

Feature selection is beneficial for improving the performance of general machine learning tasks by extracting an informative subset from the high-dimensional features. Conventional feature selection methods usually ignore the class imbalance problem, thus the selected features will be biased towards the majority class. Considering that F-measure is a more reasonable performance measure than accuracy for imbalanced data, this paper presents an effective feature selection algorithm that explores the class imbalance issue by optimizing F-measures. Since F-measure optimization can be decomposed into a series of cost-sensitive classification problems, we investigate the cost-sensitive feature selection by generating and assigning different costs to each class with rigorous theory guidance. After solving a series of cost-sensitive feature selection problems, features corresponding to the best F-measure will be selected. In this way, the selected features will fully represent the properties of all classes. Experimental results on popular benchmarks and challenging real-world data sets demonstrate the significance of cost-sensitive feature selection for the imbalanced data setting and validate the effectiveness of the proposed method.

cs.CV↗

arXiv2019

Assessment of central serous chorioretinopathy (CSC) depicted on color fundus photographs using deep Learning

To investigate whether and to what extent central serous chorioretinopathy (CSC) depicted on color fundus photographs can be assessed using deep learning technology. We collected a total of 2,504 fundus images acquired on different subjects. We verified the CSC status of these images using their corresponding optical coherence tomography (OCT) images. A total of 1,329 images depicted CSC. These images were preprocessed and normalized. This resulting dataset was randomly split into three parts in the ratio of 8:1:1 respectively for training, validation, and testing purposes. We used the deep learning architecture termed InceptionV3 to train the classifier. We performed nonparametric receiver operating characteristic (ROC) analyses to assess the capability of the developed algorithm to identify CSC. The Kappa coefficient between the two raters was 0.48 (p < 0.001), while the Kappa coefficients between the computer and the two raters were 0.59 (p < 0.001) and 0.33 (p < 0.05).Our experiments showed that the computer algorithm based on deep learning can assess CSC depicted on color fundus photographs in a relatively reliable and consistent way.

cs.CV↗

arXiv2018

Cycle Ramsey numbers for random graphs

Let $C_{n}$ be a cycle of length $n$. As an application of Szemerédi's regularity lemma, Łuczak ($R(C_n,C_n,C_n)\leq (4+o(1))n$, J. Combin. Theory Ser. B, 75 (1999), 174--187) in fact established that $K_{(8+o(1))n}\to(C_{2n+1},C_{2n+1},C_{2n+1})$. In this paper, we strengthen several results involving cycles. Let $\mathcal{G}(n,p)$ be the random graph. We prove that for fixed $0 0$, there exists an integer $n_0$ such that for all integer $n_3>n_0$, we have a.a.s. that \begin{align*} \mathcal{G}((8+δ)n_1,p) \to (C_{2n_1+1},C_{2n_2+1},C_{2n_3+1}). \end{align*} Moreover, we prove that for fixed $0 0$ with same order, i.e. $n_2=Θ(n_1)$ and $n_3=Θ(n_1)$, we have a.a.s. that \begin{align*} \mathcal{G}(2n_1+n_2+n_3+o(1)n_1,p) \to (C_{2n_1},C_{2n_2},C_{2n_3}). \end{align*} Similar results for the two color case are also obtained.

math.CO↗

arXiv2017

Unveiling soliton booting dynamics in ultrafast fiber lasers

Ultrafast fiber lasers play important roles in many aspects of our life for their various applications in fields ranging from fundamental sciences to industrial purposes. Passive mode-locking technique is a key step to realize the ultrafast soliton fiber lasers. However, the booting dynamics of the soliton fiber lasers have not yet been well understood. Herein, we reveal the soliton buildup dynamics of ultrafast fiber lasers operating both in anomalous and net-normal dispersion regimes. Based on the advanced experimental methodologies of spatio-temporal reconstruction and dispersive Fourier transform (DFT), the soliton booting dynamics are analyzed in the time and spectral domains. It was found that the booting dynamics of conventional and dissipative solitons operating in the anomalous and net-normal dispersion regimes, respectively, are different from each other due to the different pulse shaping mechanisms. In particular, the spectral interference pattern with strong relaxation oscillation behavior was observed near the mode-locking transition for conventional soliton, while no relaxation oscillation of spectral pattern was obtained for dissipative soliton. We firstly revealed that the spectral pattern distributions are induced by the transient structured soliton formation during the pulse shaping from the noise background. The experimental results were verified by the theoretical simulations. The obtained results would provide a general guideline for understanding the soliton booting dynamics in ultrafast fiber lasers, and will prove to be fruitful to the various communities interested in solitons and fiber lasers.

physics.optics↗

arXiv2015

Microfiber-based few-layer black phosphorus saturable absorber for ultra-fast fiber laser

Few-layer black phosphorus (BP), as the most alluring graphene analogue owing to its similar structure as graphene and thickness dependent direct band-gap, has now triggered a new wave of research on two-dimensional (2D) materials based photonics and optoelectronics. However, a major obstacle of practical applications for few-layer BPs comes from their instabilities of laser-induced optical damage. Herein, we demonstrate that, few-layer BPs, fabricated through the liquid exfoliation approach, can be developed as a new and practical saturable absorber (SA) by depositing few-layer BPs with microfiber. The saturable absorption property of few-layer BPs had been verified through an open-aperture z-scan measurement at the telecommunication band and the microfiber-based BP device had been found to show a saturable average power of ~4.5 mW and a modulation depth of 10.9%, which is further confirmed through a balanced twin detection measurement. By further integrating this optical SA device into an erbium-doped fiber laser, it was found that it can deliver the mode-locked pulse with duration down to 940 fs with central wavelength tunable from 1532 nm to 1570 nm. The prevention of BP from oxidation through the 'lateral interaction scheme' owing to this microfiber-based few-layer BP SA device might partially mitigate the optical damage problem of BP. Our results not only demonstrate that black phosphorus might be another promising SA material for ultrafast photonics, but also provide a practical solution to solve the optical damage problem of black phosphorus by assembling with waveguide structures such as microfiber.

physics.optics↗

arXiv2014

Femtosecond pulse generation from a Topological Insulator mode-locked fiber laser

We reported on the generation of femtosecond pulse in an anomalous-dispersion fiber ring laser by using a polyvinyl alcohol (PVA)-based Topological Insulator (TI), Bi2Se3 saturable absorber (SA). The PVA-TI composite has a low saturable optical intensity of 12 MW/cm2 and a modulation depth of ~3.9%. By incorporating the fabricated PVA-TISA into a fiber laser, mode-locking operation could be achieved at a low pump threshold of 25 mW. After an optimization of the cavity parameters, optical pulse with ~660 fs centered at 1557.5 nm wavelength had been generated. The experimental results demonstrate that the PVA could be an excellent host material for fabricating high-performance TISA, and also indicate that the filmy PVA-TISA is indeed a good candidate for ultrafast saturable absorption device.

physics.optics↗