arXiv subjects

Jun Wu

Publications and source records attributed to Jun Wu.

At least 199 records · Page 11Linked to original sources

arXiv2020

GAN-based Recommendation with Positive-Unlabeled Sampling

Recommender systems are popular tools for information retrieval tasks on a large variety of web applications and personalized products. In this work, we propose a Generative Adversarial Network based recommendation framework using a positive-unlabeled sampling strategy. Specifically, we utilize the generator to learn the continuous distribution of user-item tuples and design the discriminator to be a binary classifier that outputs the relevance score between each user and each item. Meanwhile, positive-unlabeled sampling is applied in the learning procedure of the discriminator. Theoretical bounds regarding positive-unlabeled sampling and optimalities of convergence for the discriminators and the generators are provided. We show the effectiveness and efficiency of our framework on three publicly accessible data sets with eight ranking-based evaluation metrics in comparison with thirteen popular baselines.

cs.IR↗

arXiv2020

Achievable Rate Analysis and Phase Shift Optimization on Intelligent Reflecting Surface with Hardware Impairments

Intelligent reflecting surface (IRS) is envisioned as a promising hardware solution to hardware cost and energy consumption in the fifth-generation (5G) mobile communication network. It exhibits great advantages in enhancing data transmission, but may suffer from performance degradation caused by inherent hardware impairment (HWI). For analysing the achievable rate (ACR) and optimizing the phase shifts in the IRS-aided wireless communication system with HWI, we consider that the HWI appears at both the IRS and the signal transceivers. On this foundation, first, we derive the closed-form expression of the average ACR and the IRS utility. Then, we formulate optimization problems to optimize the IRS phase shifts by maximizing the signal-to-noise ratio (SNR) at the receiver side, and obtain the solution by transforming non-convex problems into semidefinite programming (SDP) problems. Subsequently, we compare the IRS with the conventional decode-and-forward (DF) relay in terms of the ACR and the utility. Finally, we carry out simulations to verify the theoretical analysis, and evaluate the impact of the channel estimation errors and residual phase noises on the optimization performance. Our results reveal that the HWI reduces the ACR and the IRS utility, and begets more serious performance degradation with more reflecting elements. Although the HWI has an impact on the IRS, it still leaves opportunities for the IRS to surpass the conventional DF relay, when the number of reflecting elements is large enough or the transmitting power is sufficiently high.

cs.IT↗

arXiv2020

Structure Learning for Cyclic Linear Causal Models

We consider the problem of structure learning for linear causal models based on observational data. We treat models given by possibly cyclic mixed graphs, which allow for feedback loops and effects of latent confounders. Generalizing related work on bow-free acyclic graphs, we assume that the underlying graph is simple. This entails that any two observed variables can be related through at most one direct causal effect and that (confounding-induced) correlation between error terms in structural equations occurs only in absence of direct causal effects. We show that, despite new subtleties in the cyclic case, the considered simple cyclic models are of expected dimension and that a previously considered criterion for distributional equivalence of bow-free acyclic graphs has an analogue in the cyclic case. Our result on model dimension justifies in particular score-based methods for structure learning of linear Gaussian mixed graph models, which we implement via greedy search.

math.ST↗

arXiv2020

Continuous Transfer Learning with Label-informed Distribution Alignment

Transfer learning has been successfully applied across many high-impact applications. However, most existing work focuses on the static transfer learning setting, and very little is devoted to modeling the time evolving target domain, such as the online reviews for movies. To bridge this gap, in this paper, we study a novel continuous transfer learning setting with a time evolving target domain. One major challenge associated with continuous transfer learning is the potential occurrence of negative transfer as the target domain evolves over time. To address this challenge, we propose a novel label-informed C-divergence between the source and target domains in order to measure the shift of data distributions as well as to identify potential negative transfer. We then derive the error bound for the target domain using the empirical estimate of our proposed C-divergence. Furthermore, we propose a generic adversarial Variational Auto-encoder framework named TransLATE by minimizing the classification error and C-divergence of the target domain between consecutive time stamps in a latent feature space. In addition, we define a transfer signature for characterizing the negative transfer based on C-divergence, which indicates that larger C-divergence implies a higher probability of negative transfer in real scenarios. Extensive experiments on synthetic and real data sets demonstrate the effectiveness of our TransLATE framework.

cs.LG↗

arXiv2020

A framework for adaptive width control of dense contour-parallel toolpaths in fused deposition modeling

3D printing techniques such as Fused Deposition Modeling (FDM) have enabled the fabrication of complex geometry quickly and cheaply. High stiffness parts are produced by filling the 2D polygons of consecutive layers with contour-parallel extrusion toolpaths. Uniform width toolpaths consisting of inward offsets from the outline polygons produce over- and underfill regions in the center of the shape, which are especially detrimental to the mechanical performance of thin parts. In order to fill shapes with arbitrary diameter densely the toolpaths require adaptive width. Existing approaches for generating toolpaths with adaptive width result in a large variation in widths, which for some hardware systems is difficult to realize accurately. In this paper we present a framework which supports multiple schemes to generate toolpaths with adaptive width, by employing a function to decide the number of beads and their widths. Furthermore, we propose a novel scheme which reduces extreme bead widths, while limiting the number of altered toolpaths. We statistically validate the effectiveness of our framework and this novel scheme on a data set of representative 3D models, and physically validate it by developing a technique, called back pressure compensation, for off-the-shelf FDM systems to effectively realize adaptive width.

cs.GR↗

arXiv2020

Adaptive Distributed Laser Charging for Efficient Wireless Power Transfer

Distributed laser charging (DLC) is a wireless power transfer technology for mobile electronics. Similar to traditional wireless charging systems, the DLC system can only provide constant power to charge a battery. However, Li-ion battery needs dynamic input current and voltage, thus power, in order to optimize battery charging performance. Therefore, neither power transmission efficiency nor battery charging performance can be optimized by the DLC system. We at first propose an adaptive DLC (ADLC) system to optimize wireless power transfer efficiency and battery charging performance. Then, we analyze ADLC's power conversion to depict the adaptation mechanism. Finally, we evaluate the ADLC's power conversion performance by simulation, which illustrates its efficiency improvement by saving at least 60.4% of energy, comparing with the fixed-power charging system.

eess.SY↗

arXiv2020

Progressive DARTS: Bridging the Optimization Gap for NAS in the Wild

With the rapid development of neural architecture search (NAS), researchers found powerful network architectures for a wide range of vision tasks. However, it remains unclear if the searched architecture can transfer across different types of tasks as manually designed ones did. This paper puts forward this problem, referred to as NAS in the wild, which explores the possibility of finding the optimal architecture in a proxy dataset and then deploying it to mostly unseen scenarios. We instantiate this setting using a currently popular algorithm named differentiable architecture search (DARTS), which often suffers unsatisfying performance while being transferred across different tasks. We argue that the accuracy drop originates from the formulation that uses a super-network for search but a sub-network for re-training. The different properties of these stages have resulted in a significant optimization gap, and consequently, the architectural parameters "over-fit" the super-network. To alleviate the gap, we present a progressive method that gradually increases the network depth during the search stage, which leads to the Progressive DARTS (P-DARTS) algorithm. With a reduced search cost (7 hours on a single GPU), P-DARTS achieves improved performance on both the proxy dataset (CIFAR10) and a few target problems (ImageNet classification, COCO detection and three ReID benchmarks). Our code is available at \url{https://github.com/chenxin061/pdarts}.

cs.CV↗

arXiv2020

A PXI-based Multi-channel Data Acquisition System for Fast Transient Pulses

In this paper, we design a PXI-based, multi-channel data-acquisition system (DAS) mainly applicable to recording one-shot fast transient pulses in nuclear physics experiments. The system consists of one NI PXIe-1085 chassis, containing a controller card and at most 16 data-acquisition (DAQ) cards. Every single DAQ card has a sampling rate of 1GS/s and a 12bit vertical resolution with the PXI interface and SFP+ transceiver for data transmission. When the system is put into operation near the pulsed radiation source, the SFP+ optical fiber channel enables a timely data transmission to a remote server. All of these cards in the chassis can be synchronized using PXI timing and triggering resources. Additionally, a simple DAS software is developed to display the pulsed signals captured and communicate with the host PC for remote control and data upload. After careful calibration, preliminary tests show that every DAQ channel achieves an analog bandwidth higher than 200MHz and an ENOB of more than 9 bits at a 1GS/s sampling rate. Owing to such high speed and resolution, the system may facilitate improvements in extracting maximum information from transient signals. Furthermore, with great scalability and high-speed data transmission, the system can be used for other nuclear physics experiments.

physics.ins-det↗

arXiv2019

Robust Data Preprocessing for Machine-Learning-Based Disk Failure Prediction in Cloud Production Environments

To provide proactive fault tolerance for modern cloud data centers, extensive studies have proposed machine learning (ML) approaches to predict imminent disk failures for early remedy and evaluated their approaches directly on public datasets (e.g., Backblaze SMART logs). However, in real-world production environments, the data quality is imperfect (e.g., inaccurate labeling, missing data samples, and complex failure types), thereby degrading the prediction accuracy. We present RODMAN, a robust data preprocessing pipeline that refines data samples before feeding them into ML models. We start with a large-scale trace-driven study of over three million disks from Alibaba Cloud's data centers, and motivate the practical challenges in ML-based disk failure prediction. We then design RODMAN with three data preprocessing echniques, namely failure-type filtering, spline-based data filling, and automated pre-failure backtracking, that are applicable for general ML models. Evaluation on both the Alibaba and Backblaze datasets shows that RODMAN improves the prediction accuracy compared to without data preprocessing under various settings.

cs.LG↗

arXiv2019

Multi-reflection model of subwavelength grating diffraction based on simplified modal method

A multi-reflection model of grating diffraction based on the simplified modal method is proposed. Simulation results for a guided mode resonance Brewster grating using our method and rigorous coupled wave analysis are presented to verify our model. The solution of our method is in good agreement with that of rigorous coupled wave analysis. Benefiting from its clear physical view, this model helps us to better understand the diffraction process inside subwavelength gratings. On the basis of the multi-reflection model, we developed a matrix Fabry-Perot (FP) resonance condition and a single-mode resonance condition to evaluate the resonance wavelength. These resonance conditions may be helpful for simplifying design of guided mode resonance (GMR) gratings.

physics.optics↗

arXiv2019

Hierarchical Attention Networks for Medical Image Segmentation

The medical image is characterized by the inter-class indistinction, high variability, and noise, where the recognition of pixels is challenging. Unlike previous self-attention based methods that capture context information from one level, we reformulate the self-attention mechanism from the view of the high-order graph and propose a novel method, namely Hierarchical Attention Network (HANet), to address the problem of medical image segmentation. Concretely, an HA module embedded in the HANet captures context information from neighbors of multiple levels, where these neighbors are extracted from the high-order graph. In the high-order graph, there will be an edge between two nodes only if the correlation between them is high enough, which naturally reduces the noisy attention information caused by the inter-class indistinction. The proposed HA module is robust to the variance of input and can be flexibly inserted into the existing convolution neural networks. We conduct experiments on three medical image segmentation tasks including optic disc/cup segmentation, blood vessel segmentation, and lung segmentation. Extensive results show our method is more effective and robust than the existing state-of-the-art methods.

cs.CV↗

arXiv2019

A generalized perfect vortex beam with controllable impulse ring profile

Perfect optical vortices (POVs) provide an enabling solution to address the predicament induced by the strong dependence of classical optical vortices on theirs carried topological charges. Here, a type of generalized POVs with controllable impulse ring profile was proposed and demonstrated. Especially, a type of "absolute" dark POVs surrounded by two bright ringlobes in each side was presented, which provides a perfect annular potential well along those dark impulse rings for trapping steadily low-index particles, cells, or quantum gas, etc. In further, several POVs with different impulse ring profiles, including conventional POVs with bright rings, dark POVs mentioned above, and also POVs with controllable impulse ring profile, were demonstrated. This work opens up new possibilities to reshape arbitrarily the impulse ring profile for perfect vortices, and this type of novel POVs will enrich functions of optical vortices and it should be of high interest for its potential applications in optical manipulation, both quantum and classical optical communications, enhanced optical imaging, and also novel structured pumping lasers, etc..

physics.optics↗

arXiv2019

CrossFill: Foam Structures with Graded Density for Continuous Material Extrusion

The fabrication flexibility of 3D printing has sparked a lot of interest in designing structures with spatially graded material properties. In this paper, we propose a new type of density graded structure that is particularly designed for 3D printing systems based on filament extrusion. In order to ensure high-quality fabrication results, extrusion-based 3D printing requires not only that the structures are self-supporting, but also that extrusion toolpaths are continuous and free of self-overlap. The structure proposed in this paper, called CrossFill, complies with these requirements. In particular, CrossFill is a self-supporting foam structure, for which each layer is fabricated by a single, continuous and overlap-free path of material extrusion. Our method for generating CrossFill is based on a space-filling surface that employs spatially varying subdivision levels. Dithering of the subdivision levels is performed to accurately reproduce a prescribed density distribution. We demonstrate the effectiveness of CrossFill on a number of experimental tests and applications.

cs.CG↗

arXiv2019

DEMO-Net: Degree-specific Graph Neural Networks for Node and Graph Classification

Graph data widely exist in many high-impact applications. Inspired by the success of deep learning in grid-structured data, graph neural network models have been proposed to learn powerful node-level or graph-level representation. However, most of the existing graph neural networks suffer from the following limitations: (1) there is limited analysis regarding the graph convolution properties, such as seed-oriented, degree-aware and order-free; (2) the node's degree-specific graph structure is not explicitly expressed in graph convolution for distinguishing structure-aware node neighborhoods; (3) the theoretical explanation regarding the graph-level pooling schemes is unclear. To address these problems, we propose a generic degree-specific graph neural network named DEMO-Net motivated by Weisfeiler-Lehman graph isomorphism test that recursively identifies 1-hop neighborhood structures. In order to explicitly capture the graph topology integrated with node attributes, we argue that graph convolution should have three properties: seed-oriented, degree-aware, order-free. To this end, we propose multi-task graph convolution where each task represents node representation learning for nodes with a specific degree value, thus leading to preserving the degree-specific graph structure. In particular, we design two multi-task learning methods: degree-specific weight and hashing functions for graph convolution. In addition, we propose a novel graph-level pooling/readout scheme for learning graph representation provably lying in a degree-specific Hilbert kernel space. The experimental results on several node and graph classification benchmark data sets demonstrate the effectiveness and efficiency of our proposed DEMO-Net over state-of-the-art graph neural network models.

cs.LG↗

arXiv2019

Design and Optimization of Conforming Lattice Structures

Inspired by natural cellular materials such as trabecular bone, lattice structures have been developed as a new type of lightweight material. In this paper we present a novel method to design lattice structures that conform with both the principal stress directions and the boundary of the optimized shape. Our method consists of two major steps: the first optimizes concurrently the shape (including its topology) and the distribution of orthotropic lattice materials inside the shape to maximize stiffness under application-specific external loads; the second takes the optimized configuration (i.e. locally-defined orientation, porosity, and anisotropy) of lattice materials from the previous step, and extracts a globally consistent lattice structure by field-aligned parameterization. Our approach is robust and works for both 2D planar and 3D volumetric domains. Numerical results and physical verifications demonstrate remarkable structural properties of conforming lattice structures generated by our method.

cs.CE↗

arXiv2019

Progressive Differentiable Architecture Search: Bridging the Depth Gap between Search and Evaluation

Recently, differentiable search methods have made major progress in reducing the computational costs of neural architecture search. However, these approaches often report lower accuracy in evaluating the searched architecture or transferring it to another dataset. This is arguably due to the large gap between the architecture depths in search and evaluation scenarios. In this paper, we present an efficient algorithm which allows the depth of searched architectures to grow gradually during the training procedure. This brings two issues, namely, heavier computational overheads and weaker search stability, which we solve using search space approximation and regularization, respectively. With a significantly reduced search time (~7 hours on a single GPU), our approach achieves state-of-the-art performance on both the proxy dataset (CIFAR10 or CIFAR100) and the target dataset (ImageNet). Code is available at https://github.com/chenxin061/pdarts.

cs.CV↗

arXiv2019

Market Manipulation of Bitcoin: Evidence from Mining the Mt. Gox Transaction Network

The cryptocurrency market is a very huge market without effective supervision. It is of great importance for investors and regulators to recognize whether there are market manipulation and its manipulation patterns. This paper proposes an approach to mine the transaction networks of exchanges for answering this question.By taking the leaked transaction history of Mt. Gox Bitcoin exchange as a sample,we first divide the accounts into three categories according to its characteristic and then construct the transaction history into three graphs. Many observations and findings are obtained via analyzing the constructed graphs. To evaluate the influence of the accounts' transaction behavior on the Bitcoin exchange price,the graphs are reconstructed into series and reshaped as matrices. By using singular value decomposition (SVD) on the matrices, we identify many base networks which have a great correlation with the price fluctuation. When further analyzing the most important accounts in the base networks, plenty of market manipulation patterns are found. According to these findings, we conclude that there was serious market manipulation in Mt. Gox exchange and the cryptocurrency market must strengthen the supervision.

q-fin.GN↗

arXiv2018

Autoencoder Based Residual Deep Networks for Robust Regression Prediction and Spatiotemporal Estimation

To have a superior generalization, a deep learning neural network often involves a large size of training sample. With increase of hidden layers in order to increase learning ability, neural network has potential degradation in accuracy. Both could seriously limit applicability of deep learning in some domains particularly involving predictions of continuous variables with a small size of samples. Inspired by residual convolutional neural network in computer vision and recent findings of crucial shortcuts in the brains in neuroscience, we propose an autoencoder-based residual deep network for robust prediction. In a nested way, we leverage shortcut connections to implement residual mapping with a balanced structure for efficient propagation of error signals. The novel method is demonstrated by multiple datasets, imputation of high spatiotemporal resolution non-randomness missing values of aerosol optical depth, and spatiotemporal estimation of fine particulate matter <2.5 μm, achieving the cutting edge of accuracy and efficiency. Our approach is also a general-purpose regression learner to be applicable in diverse domains.

cs.LG↗