arXiv subjects

Shuo Wang

Publications and source records attributed to Shuo Wang.

At least 559 records · Page 31Linked to original sources

arXiv2019

Automatic Brain Tumour Segmentation and Biophysics-Guided Survival Prediction

Gliomas are the most common malignant brain tumourswith intrinsic heterogeneity. Accurate segmentation of gliomas and theirsub-regions on multi-parametric magnetic resonance images (mpMRI)is of great clinical importance, which defines tumour size, shape andappearance and provides abundant information for preoperative diag-nosis, treatment planning and survival prediction. Recent developmentson deep learning have significantly improved the performance of auto-mated medical image segmentation. In this paper, we compare severalstate-of-the-art convolutional neural network models for brain tumourimage segmentation. Based on the ensembled segmentation, we presenta biophysics-guided prognostic model for patient overall survival predic-tion which outperforms a data-driven radiomics approach. Our methodwon the second place of the MICCAI 2019 BraTS Challenge for theoverall survival prediction.

eess.IV↗

arXiv2019

Time-Aware Gated Recurrent Unit Networks for Road Surface Friction Prediction Using Historical Data

An accurate road surface friction prediction algorithm can enable intelligent transportation systems to share timely road surface condition to the public for increasing the safety of the road users. Previously, scholars developed multiple prediction models for forecasting road surface conditions using historical data. However, road surface condition data cannot be perfectly collected at every timestamp, e.g. the data collected by on-vehicle sensors may be influenced when vehicles cannot travel due to economic cost issue or weather issues. Such resulted missing values in the collected data can damage the effectiveness and accuracy of the existing prediction methods since they are assumed to have the input data with a fixed temporal resolution. This study proposed a road surface friction prediction model employing a Gated Recurrent Unit network-based decay mechanism (GRU-D) to handle the missing values. The evaluation results present that the proposed GRU-D networks outperform all baseline models. The impact of missing rate on predictive accuracy, learning efficiency and learned decay rate are analyzed as well. The findings can help improve the prediction accuracy and efficiency of forecasting road surface friction using historical data sets with missing values, therefore mitigating the impact of wet or icy road conditions on traffic safety.

cs.LG↗

arXiv2019

Road Surface Friction Prediction Using Long Short-Term Memory Neural Network Based on Historical Data

Road surface friction significantly impacts traffic safety and mobility. A precise road surface friction prediction model can help to alleviate the influence of inclement road conditions on traffic safety, Level of Service, traffic mobility, fuel efficiency, and sustained economic productivity. Most related previous studies are laboratory-based methods that are difficult for practical implementation. Moreover, in other data-driven methods, the demonstrated time-series features of road surface conditions have not been considered. This study employed a Long-Short Term Memory (LSTM) neural network to develop a data-driven road surface friction prediction model based on historical data. The proposed prediction model outperformed the other baseline models in terms of the lowest value of predictive performance measurements. The influence of the number of time-lags and the predicting time interval on predictive accuracy was analyzed. In addition, the influence of adding road surface water thickness, road surface temperature and air temperature on predictive accuracy also were investigated. The findings of this study can support road maintenance strategy development and decision making, thus mitigating the impact of inclement road conditions on traffic mobility and safety. Future work includes a modified LSTM-based prediction model development by accommodating flexible time intervals between time-lags.

eess.SP↗

arXiv2019

Tutorial: Complexity analysis of Singular Value Decomposition and its variants

We compared the regular Singular Value Decomposition (SVD), truncated SVD, Krylov method and Randomized PCA, in terms of time and space complexity. It is well-known that Krylov method and Randomized PCA only performs well when k << n, i.e. the number of eigenpair needed is far less than that of matrix size. We compared them for calculating all the eigenpairs. We also discussed the relationship between Principal Component Analysis and SVD.

math.NA↗

arXiv2019

REQ-YOLO: A Resource-Aware, Efficient Quantization Framework for Object Detection on FPGAs

Deep neural networks (DNNs), as the basis of object detection, will play a key role in the development of future autonomous systems with full autonomy. The autonomous systems have special requirements of real-time, energy-efficient implementations of DNNs on a power-constrained system. Two research thrusts are dedicated to performance and energy efficiency enhancement of the inference phase of DNNs. The first one is model compression techniques while the second is efficient hardware implementation. Recent works on extremely-low-bit CNNs such as the binary neural network (BNN) and XNOR-Net replace the traditional floating-point operations with binary bit operations which significantly reduces the memory bandwidth and storage requirement. However, it suffers from non-negligible accuracy loss and underutilized digital signal processing (DSP) blocks of FPGAs. To overcome these limitations, this paper proposes REQ-YOLO, a resource-aware, systematic weight quantization framework for object detection, considering both algorithm and hardware resource aspects in object detection. We adopt the block-circulant matrix method and propose a heterogeneous weight quantization using the Alternating Direction Method of Multipliers (ADMM), an effective optimization technique for general, non-convex optimization problems. To achieve real-time, highly-efficient implementations on FPGA, we present the detailed hardware implementation of block circulant matrices on CONV layers and develop an efficient processing element (PE) structure supporting the heterogeneous weight quantization, CONV dataflow and pipelining techniques, design optimization, and a template-based automatic synthesis framework to optimally exploit hardware resource. Experimental results show that our proposed REQ-YOLO framework can significantly compress the YOLO model while introducing very small accuracy degradation.

cs.CV↗

arXiv2019

Improving Back-Translation with Uncertainty-based Confidence Estimation

While back-translation is simple and effective in exploiting abundant monolingual corpora to improve low-resource neural machine translation (NMT), the synthetic bilingual corpora generated by NMT models trained on limited authentic bilingual data are inevitably noisy. In this work, we propose to quantify the confidence of NMT model predictions based on model uncertainty. With word- and sentence-level confidence measures based on uncertainty, it is possible for back-translation to better cope with noise in synthetic bilingual corpora. Experiments on Chinese-English and English-German translation tasks show that uncertainty-based confidence estimation significantly improves the performance of back-translation.

cs.CL↗

arXiv2019

Predicting charge density distribution of materials using a local-environment-based graph convolutional network

Electron charge density distribution of materials is one of the key quantities in computational materials science as theoretically it determines the ground state energy and practically it is used in many materials analyses. However, the scaling of density functional theory calculations with number of atoms limits the usage of charge-density-based calculations and analyses. Here we introduce a machine learning scheme with local-environment-based graphs and graph convolutional neural networks to predict charge density on grid-points from crystal structure. We show the accuracy of this scheme through a comparison of predicted charge densities as well as properties derived from the charge density, and the scaling is O(N). More importantly, the transferability is shown to be high with respect to different compositions and structures, which results from the explicit encoding of geometry.

physics.comp-ph↗

arXiv2019

A General Deep Learning Framework for Network Reconstruction and Dynamics Learning

Many complex processes can be viewed as dynamical systems on networks. However, in real cases, only the performances of the system are known, the network structure and the dynamical rules are not observed. Therefore, recovering latent network structure and dynamics from observed time series data are important tasks because it may help us to open the black box, and even to build up the model of a complex system automatically. Although this problem hosts a wealth of potential applications in biology, earth science, and epidemics etc., conventional methods have limitations. In this work, we introduce a new framework, Gumbel Graph Network (GGN), which is a model-free, data-driven deep learning framework to accomplish the reconstruction of both network connections and the dynamics on it. Our model consists of two jointly trained parts: a network generator that generating a discrete network with the Gumbel Softmax technique; and a dynamics learner that utilizing the generated network and one-step trajectory value to predict the states in future steps. We exhibit the universality of our framework on different kinds of time-series data: with the same structure, our model can be trained to accurately recover the network structure and predict future states on continuous, discrete, and binary dynamics, and outperforms competing network reconstruction methods.

cond-mat.dis-nn↗

arXiv2019

A support vector regression-based multi-fidelity surrogate model

Computational simulations with different fidelity have been widely used in engineering design. A high-fidelity (HF) model is generally more accurate but also more time-consuming than an low-fidelity (LF) model. To take advantages of both HF and LF models, multi-fidelity surrogate models that aim to integrate information from both HF and LF models have gained increasing popularity. In this paper, a multi-fidelity surrogate model based on support vector regression named as Co_SVR is developed by combining HF and LF models. In Co_SVR, a kernel function is used to map the map the difference between the HF and LF models. Besides, a heuristic algorithm is used to obtain the optimal parameters of Co_SVR. The proposed Co_SVR is compared with two popular multi-fidelity surrogate models Co_Kriging model, Co_RBF model, and their single-fidelity surrogates through several numerical cases and a pressure vessel design problem. The results show that Co_SVR provides competitive prediction accuracy for numerical cases, and presents a better performance compared with the Co_Kriging and Co_RBF models and single-fidelity surrogate models.

stat.ML↗

arXiv2019

Experimental investigation of quantum key distribution over water channel

Quantum key distribution (QKD) has undergone significant development in recent decades, particularly with respect to free-space (air) and optical fiber channels. Here, we report the first proof-of-principle experiment for the BB84 protocol QKD over a water channel. Firstly, we demonstrate again the polarization preservation properties of the water channel in optical transmission according to the measured Mueller matrix, which is close to the unit matrix. The reason for the polarization preservation, revealed by Monte Carlo simulation, is that almost all the received photons are unscattered.Then, we performed the first polarization encoding BB84 protocol QKD over a 2.37m water channel. The results show that QKD can be performed with a low quantum bit error rate (QBER), less than 3.5%, with different attenuation coefficients.

quant-ph↗

arXiv2019

CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identification

Urban traffic optimization using traffic cameras as sensors is driving the need to advance state-of-the-art multi-target multi-camera (MTMC) tracking. This work introduces CityFlow, a city-scale traffic camera dataset consisting of more than 3 hours of synchronized HD videos from 40 cameras across 10 intersections, with the longest distance between two simultaneous cameras being 2.5 km. To the best of our knowledge, CityFlow is the largest-scale dataset in terms of spatial coverage and the number of cameras/videos in an urban environment. The dataset contains more than 200K annotated bounding boxes covering a wide range of scenes, viewing angles, vehicle models, and urban traffic flow conditions. Camera geometry and calibration information are provided to aid spatio-temporal analysis. In addition, a subset of the benchmark is made available for the task of image-based vehicle re-identification (ReID). We conducted an extensive experimental evaluation of baselines/state-of-the-art approaches in MTMC tracking, multi-target single-camera (MTSC) tracking, object detection, and image-based ReID on this dataset, analyzing the impact of different network architectures, loss functions, spatio-temporal models and their combinations on task effectiveness. An evaluation server is launched with the release of our benchmark at the 2019 AI City Challenge (https://www.aicitychallenge.org/) that allows researchers to compare the performance of their newest techniques. We expect this dataset to catalyze research in this field, propel the state-of-the-art forward, and lead to deployed traffic optimization(s) in the real world.

cs.CV↗

arXiv2019

Hindsight Generative Adversarial Imitation Learning

Compared to reinforcement learning, imitation learning (IL) is a powerful paradigm for training agents to learn control policies efficiently from expert demonstrations. However, in most cases, obtaining demonstration data is costly and laborious, which poses a significant challenge in some scenarios. A promising alternative is to train agent learning skills via imitation learning without expert demonstrations, which, to some extent, would extremely expand imitation learning areas. To achieve such expectation, in this paper, we propose Hindsight Generative Adversarial Imitation Learning (HGAIL) algorithm, with the aim of achieving imitation learning satisfying no need of demonstrations. Combining hindsight idea with the generative adversarial imitation learning (GAIL) framework, we realize implementing imitation learning successfully in cases of expert demonstration data are not available. Experiments show that the proposed method can train policies showing comparable performance to current imitation learning methods. Further more, HGAIL essentially endows curriculum learning mechanism which is critical for learning policies.

cs.LG↗

arXiv2019

Improved Selective Refinement Network for Face Detection

As a long-standing problem in computer vision, face detection has attracted much attention in recent decades for its practical applications. With the availability of face detection benchmark WIDER FACE dataset, much of the progresses have been made by various algorithms in recent years. Among them, the Selective Refinement Network (SRN) face detector introduces the two-step classification and regression operations selectively into an anchor-based face detector to reduce false positives and improve location accuracy simultaneously. Moreover, it designs a receptive field enhancement block to provide more diverse receptive field. In this report, to further improve the performance of SRN, we exploit some existing techniques via extensive experiments, including new data augmentation strategy, improved backbone network, MS COCO pretraining, decoupled classification module, segmentation branch and Squeeze-and-Excitation block. Some of these techniques bring performance improvements, while few of them do not well adapt to our baseline. As a consequence, we present an improved SRN face detector by combining these useful techniques together and obtain the best performance on widely used face detection benchmark WIDER FACE dataset.

cs.CV↗

arXiv2018

Support Vector Guided Softmax Loss for Face Recognition

Face recognition has witnessed significant progresses due to the advances of deep convolutional neural networks (CNNs), the central challenge of which, is feature discrimination. To address it, one group tries to exploit mining-based strategies (\textit{e.g.}, hard example mining and focal loss) to focus on the informative examples. The other group devotes to designing margin-based loss functions (\textit{e.g.}, angular, additive and additive angular margins) to increase the feature margin from the perspective of ground truth class. Both of them have been well-verified to learn discriminative features. However, they suffer from either the ambiguity of hard examples or the lack of discriminative power of other classes. In this paper, we design a novel loss function, namely support vector guided softmax loss (SV-Softmax), which adaptively emphasizes the mis-classified points (support vectors) to guide the discriminative features learning. So the developed SV-Softmax loss is able to eliminate the ambiguity of hard examples as well as absorb the discriminative power of other classes, and thus results in more discrimiantive features. To the best of our knowledge, this is the first attempt to inherit the advantages of mining-based and margin-based losses into one framework. Experimental results on several benchmarks have demonstrated the effectiveness of our approach over state-of-the-arts.

cs.CV↗

arXiv2018

Optimization Matrix Factorization Recommendation Algorithm Based on Rating Centrality

Matrix factorization (MF) is extensively used to mine the user preference from explicit ratings in recommender systems. However, the reliability of explicit ratings is not always consistent, because many factors may affect the user's final evaluation on an item, including commercial advertising and a friend's recommendation. Therefore, mining the reliable ratings of user is critical to further improve the performance of the recommender system. In this work, we analyze the deviation degree of each rating in overall rating distribution of user and item, and propose the notion of user-based rating centrality and item-based rating centrality, respectively. Moreover, based on the rating centrality, we measure the reliability of each user rating and provide an optimized matrix factorization recommendation algorithm. Experimental results on two popular recommendation datasets reveal that our method gets better performance compared with other matrix factorization recommendation algorithms, especially on sparse datasets.

cs.IR↗

arXiv2018

A new porous metallic silicon dicarbide for highly efficient Li-ion battery anode identified by targeted structure search

Extensive efforts have been devoted to C-Si compound materials for improving the limited specific capacity of graphite anode and avoiding the huge volume change of Si anode in Li-ion battery, but not much progress has been made during the past decades. Here, for the first time we apply the targeted structure search by using Li in desired quantity as chemical template to regulate the bonding between C and Si, which makes searching more feasible for us to find a new stable phase of C2Si (labelled as T-C2Si) that can better fit the XRD data of silicon dicarbide synthesized before. Different from the conventional semiconducting silicon carbides, T-C2Si is not only metallic with high intrinsic conductivity for electrons transport, but also porous with regularly distributed channels in suitable size for Li ions experiencing a low energy barrier. T-C2Si exhibits a high specific capacity of 515 mA/g, a high average open-circuit voltage of 1.14 eV, and a low volume change of 1.6%. These parameters meet the requirements of an ideal anode material with high performance for electric vehicles. Moreover, our targeted search strategy guarantees the resulting anode material with a desirable specific capacity and a small volume change during charging /discharging, and it can be used to find new geometric configurations for other materials.

cond-mat.mtrl-sci↗

arXiv2018

Bulk and surface states carried supercurrent in ballistic Nb-Dirac semimetal Cd3As2 nanowire-Nb junctions

A three-dimensional Dirac semimetal has bulk Dirac cones in all three momentum directions and Fermi arc-like surface states, and can be converted into a Weyl semimetal by breaking time-reversal symmetry. However, the highly conductive bulk state usually hides the electronic transport from the surface state in Dirac semimetal. Here, we demonstrate the supercurrent carried by bulk and surface states in Nb-Cd3As2 nanowire-Nb short and long junctions, respectively. For the 1 micrometer long junction, the Fabry-Perot interferences induced oscillations of the critical supercurrent are observed, suggesting the ballistic transport of the surface states carried supercurrent, where the bulk states are decoherent and the topologically protected surface states still keep coherent. Moreover, a superconducting dome is observed in the long junction, which is attributed to the enhanced dephasing from the interaction between surface and bulk states as tuning gate voltage to increase the carrier density. The superconductivity of topological semimetal nanowire is promising for braiding of Majorana fermions toward topological quantum computing.

cond-mat.mes-hall↗

arXiv2018

Efficient Recurrent Neural Networks using Structured Matrices in FPGAs

Recurrent Neural Networks (RNNs) are becoming increasingly important for time series-related applications which require efficient and real-time implementations. The recent pruning based work ESE suffers from degradation of performance/energy efficiency due to the irregular network structure after pruning. We propose block-circulant matrices for weight matrix representation in RNNs, thereby achieving simultaneous model compression and acceleration. We aim to implement RNNs in FPGA with highest performance and energy efficiency, with certain accuracy requirement (negligible accuracy degradation). Experimental results on actual FPGA deployments shows that the proposed framework achieves a maximum energy efficiency improvement of 35.7$\times$ compared with ESE.

cs.LG↗