Searcharxiv⌕ Search

arXiv subjects

Yiwei Sun

Publications and source records attributed to Yiwei Sun.

34 records · Page 2Linked to original sources
arXiv2019

Joint Modeling of Local and Global Temporal Dynamics for Multivariate Time Series Forecasting with Missing Values

Multivariate time series (MTS) forecasting is widely used in various domains, such as meteorology and traffic. Due to limitations on data collection, transmission, and storage, real-world MTS data usually contains missing values, making it infeasible to apply existing MTS forecasting models such as linear regression and recurrent neural networks. Though many efforts have been devoted to this problem, most of them solely rely on local dependencies for imputing missing values, which ignores global temporal dynamics. Local dependencies/patterns would become less useful when the missing ratio is high, or the data have consecutive missing values; while exploring global patterns can alleviate such problems. Thus, jointly modeling local and global temporal dynamics is very promising for MTS forecasting with missing values. However, work in this direction is rather limited. Therefore, we study a novel problem of MTS forecasting with missing values by jointly exploring local and global temporal dynamics. We propose a new framework LGnet, which leverages memory network to explore global patterns given estimations from local perspectives. We further introduce adversarial training to enhance the modeling of global temporal distribution. Experimental results on real-world datasets show the effectiveness of LGnet for MTS forecasting with missing values and its robustness under various missing ratios.

cs.LG↗
arXiv2019

LMLFM: Longitudinal Multi-Level Factorization Machine

We consider the problem of learning predictive models from longitudinal data, consisting of irregularly repeated, sparse observations from a set of individuals over time. Such data often exhibit {\em longitudinal correlation} (LC) (correlations among observations for each individual over time), {\em cluster correlation} (CC) (correlations among individuals that have similar characteristics), or both. These correlations are often accounted for using {\em mixed effects models} that include {\em fixed effects} and {\em random effects}, where the fixed effects capture the regression parameters that are shared by all individuals, whereas random effects capture those parameters that vary across individuals. However, the current state-of-the-art methods are unable to select the most predictive fixed effects and random effects from a large number of variables, while accounting for complex correlation structure in the data and non-linear interactions among the variables. We propose Longitudinal Multi-Level Factorization Machine (LMLFM), to the best of our knowledge, the first model to address these challenges in learning predictive models from longitudinal data. We establish the convergence properties, and analyze the computational complexity, of LMLFM. We present results of experiments with both simulated and real-world longitudinal data which show that LMLFM outperforms the state-of-the-art methods in terms of predictive accuracy, variable selection ability, and scalability to data with large number of variables. The code and supplemental material is available at \url{https://github.com/junjieliang672/LMLFM}.

stat.ML↗
arXiv2019

Node Injection Attacks on Graphs via Reinforcement Learning

Real-world graph applications, such as advertisements and product recommendations make profits based on accurately classify the label of the nodes. However, in such scenarios, there are high incentives for the adversaries to attack such graph to reduce the node classification performance. Previous work on graph adversarial attacks focus on modifying existing graph structures, which is infeasible in most real-world applications. In contrast, it is more practical to inject adversarial nodes into existing graphs, which can also potentially reduce the performance of the classifier. In this paper, we study the novel node injection poisoning attacks problem which aims to poison the graph. We describe a reinforcement learning based method, namely NIPA, to sequentially modify the adversarial information of the injected nodes. We report the results of experiments using several benchmark data sets that show the superior performance of the proposed method NIPA, relative to the existing state-of-the-art methods.

cs.LG↗
arXiv2019

MEGAN: A Generative Adversarial Network for Multi-View Network Embedding

Data from many real-world applications can be naturally represented by multi-view networks where the different views encode different types of relationships (e.g., friendship, shared interests in music, etc.) between real-world individuals or entities. There is an urgent need for methods to obtain low-dimensional, information preserving and typically nonlinear embeddings of such multi-view networks. However, most of the work on multi-view learning focuses on data that lack a network structure, and most of the work on network embeddings has focused primarily on single-view networks. Against this background, we consider the multi-view network representation learning problem, i.e., the problem of constructing low-dimensional information preserving embeddings of multi-view networks. Specifically, we investigate a novel Generative Adversarial Network (GAN) framework for Multi-View Network Embedding, namely MEGAN, aimed at preserving the information from the individual network views, while accounting for connectivity across (and hence complementarity of and correlations between) different views. The results of our experiments on two real-world multi-view data sets show that the embeddings obtained using MEGAN outperform the state-of-the-art methods on node classification, link prediction and visualization tasks.

cs.SI↗
arXiv2019

Disruption of the $sp^2$ bonding by the compression of the $π$-electronic orbitals of graphene at various stacking orders

We investigate the behaviour of the $π$-electrons under compression and the effect of the stacking order of graphene layers. First we find that electrons can hardly be squeezed through the $sp^2$ network, regardless of the stacking order. The largely deformed electronic orbitals (mainly those of $π$-electrons) under compression along the $\textit{c}$-axis increase interlayer interaction between graphene layers as expected, but surprisingly in a similar way for the A-A and Bernal stacking. On the other hand, the large out-of-plane compression shifts the in-plane phonon frequencies of A-A stacked graphene layers significantly and very differently from Bernal stacked layers. We attribute these results to the $sp^2$-electrons filling the low-density central area in a carbon hexagon under compression for the A-A stacking, hence resulting in a non-monotonic change of the $sp^2$-bonding. The results strongly suggest not to ignore 3D features of a 2D material.

cond-mat.mtrl-sci↗
arXiv2019

Multi-View Network Embedding Via Graph Factorization Clustering and Co-Regularized Multi-View Agreement

Real-world social networks and digital platforms are comprised of individuals (nodes) that are linked to other individuals or entities through multiple types of relationships (links). Sub-networks of such a network based on each type of link correspond to distinct views of the underlying network. In real-world applications, each node is typically linked to only a small subset of other nodes. Hence, practical approaches to problems such as node labeling have to cope with the resulting sparse networks. While low-dimensional network embeddings offer a promising approach to this problem, most of the current network embedding methods focus primarily on single view networks. We introduce a novel multi-view network embedding (MVNE) algorithm for constructing low-dimensional node embeddings from multi-view networks. MVNE adapts and extends an approach to single view network embedding (SVNE) using graph factorization clustering (GFC) to the multi-view setting using an objective function that maximizes the agreement between views based on both the local and global structure of the underlying multi-view graph. Our experiments with several benchmark real-world single view networks show that GFC-based SVNE yields network embeddings that are competitive with or superior to those produced by the state-of-the-art single view network embedding methods when the embeddings are used for labeling unlabeled nodes in the networks. Our experiments with several multi-view networks show that MVNE substantially outperforms the single view methods on integrated view and the state-of-the-art multi-view methods. We further show that even when the goal is to predict labels of nodes within a single target view, MVNE outperforms its single-view counterpart suggesting that the MVNE is able to extract the information that is useful for labeling nodes in the target view from the all of the views.

cs.LG↗
arXiv2019

Improving Image Captioning by Leveraging Knowledge Graphs

We explore the use of a knowledge graphs, that capture general or commonsense knowledge, to augment the information extracted from images by the state-of-the-art methods for image captioning. The results of our experiments, on several benchmark data sets such as MS COCO, as measured by CIDEr-D, a performance metric for image captioning, show that the variants of the state-of-the-art methods for image captioning that make use of the information extracted from knowledge graphs can substantially outperform those that rely solely on the information extracted from images.

cs.CV↗
arXiv2019

Geometrically incompatible confinement of solids

The complex morphologies exhibited by spatially confined thin objects have long challenged human efforts to understand and manipulate them, from the representation of patterns in draped fabric in Renaissance art to current day efforts to engineer flexible sensors that conform to the human body. We introduce a theoretical principle, broadly generalizing Euler's {\emph{elastica}} -- a core concept of continuum mechanics that invokes the energetic preference of bending over straining a thin solid object and has been widely applied to classical and modern studies of beams and rods. We define a class of {\emph{geometrically incompatible confinement}} problems, whereby the topography imposed on a thin solid body is incompatible with its intrinsic ("target") metric and, as a consequence of Gauss' {\emph{Theorema Egregium}}, induces strain. Focusing on a prototypical example of a sheet attached to a spherical substrate, numerical simulations and analytical study demonstrate that the mechanics is governed by a principle, which we call the "Gauss-Euler {\emph{elastica}}". This emergent rule states that - despite the unavoidable strain in such an incompatible confinement - the ratio between the energies stored in straining and bending the solid may be arbitrarily small. The Gauss-Euler {\emph{elastica}} underlies a theoretical framework that greatly simplifies the daunting task of solving the highly nonlinear equations that describe thin solids at mechanical equilibrium. This development thus opens new possibilities for attacking a broad class of phenomena governed by the coupling of geometry and mechanics.

cond-mat.soft↗
arXiv2018

Wideband Hybrid Precoding for Next-Generation Backhaul/Fronthaul Based on mmWave FD-MIMO

Millimeter-wave (mmWave) communication is considered as an indispensable technique for the next-generation backhaul/fronthaul network thanks to its large transmission bandwidth. Especially for heterogeneous network (HetNet), the mmWave full-dimension (FD)-MIMO is exploited to establish the backhaul/fronthaul link between phantom-cell base stations (BSs) and macro-cell BSs, where an efficient precoding is prerequisite. Against this background, this paper proposes a principle component analysis (PCA)-based hybrid precoding for wideband mmWave MIMO backhaul/fronthaul channels. We first propose an optimal hybrid precoder by exploiting principal component analysis (PCA), whereby the optimal high dimensional frequency-selective precoder are projected to the low-dimensional frequency-flat precoder. Moreover, the combiner is designed by leveraging the weighted PCA, where the covariance of received signal is taken into account as weight to the optimal minimum mean square error (MMSE) fully-digital combiner for further improved performance. Simulations have confirmed that the proposed scheme outperforms conventional schemes in spectral efficiency (SE) and bit-error-rate (BER) performance.

cs.IT↗
arXiv2018

Machine Learning Based Hybrid Precoding for MmWave MIMO-OFDM with Dynamic Subarray

Hybrid precoding design can be challenging for broadband millimeter-wave (mmWave) massive MIMO due to the frequency-flat analog precoder in radio frequency (RF). Prior broadband hybrid precoding work usually focuses on fully-connected array (FCA), while seldom considers the energy-efficient partially-connected subarray (PCS) including the fixed subarray (FS) and dynamic subarray (DS). Against this background, this paper proposes a machine learning based broadband hybrid precoding for mmWave massive MIMO with DS. Specifically, we first propose an optimal hybrid precoder based on principal component analysis (PCA) for the FS, whereby the frequency-flat RF precoder for each subarray is extracted from the principle component of the optimal frequency-selective precoders for fully-digital MIMO. Moreover, we extend the PCA-based hybrid precoding to DS, where a shared agglomerative hierarchical clustering (AHC) algorithm developed from machine learning is proposed to group the DS for improved spectral efficiency (SE). Finally, we investigate the energy efficiency (EE) of the proposed scheme for both passive and active antennas. Simulations have confirmed that the proposed scheme outperforms conventional schemes in both SE and EE.

cs.IT↗
arXiv2018

Compositional Stochastic Average Gradient for Machine Learning and Related Applications

Many machine learning, statistical inference, and portfolio optimization problems require minimization of a composition of expected value functions (CEVF). Of particular interest is the finite-sum versions of such compositional optimization problems (FS-CEVF). Compositional stochastic variance reduced gradient (C-SVRG) methods that combine stochastic compositional gradient descent (SCGD) and stochastic variance reduced gradient descent (SVRG) methods are the state-of-the-art methods for FS-CEVF problems. We introduce compositional stochastic average gradient descent (C-SAG) a novel extension of the stochastic average gradient method (SAG) to minimize composition of finite-sum functions. C-SAG, like SAG, estimates gradient by incorporating memory of previous gradient information. We present theoretical analyses of C-SAG which show that C-SAG, like SAG, and C-SVRG, achieves a linear convergence rate when the objective function is strongly convex; However, C-CAG achieves lower oracle query complexity per iteration than C-SVRG. Finally, we present results of experiments showing that C-SAG converges substantially faster than full gradient (FG), as well as C-SVRG.

cs.LG↗
arXiv2017

Two-dimensional material based "field-effect CCD"

Charge-coupled device (CCD), along with the complementary metal-oxide-semiconductor (CMOS), is one of the major imaging technologies. Traditional CCDs rely on the charge transfer between potential wells, which gives them advantages of simple pixel structure, high sensitivity, and low noise. However, the serial transfer requires fabrication incompatible with CMOS, and leads to slower, more complex, and less flexible readout than random access. Here, we report a new-concept CCD called field-effect CCD (FE-CCD), which is based on the capacitive "coupling" between the semiconductor substrate and the 2D material. The strong field effect of the 2D material enables amplification, and non-destructive readout of the integrated charges in one pixel without transfer. We demonstrated the broadband response of our CCD (up to 1870 nm) and the successful integration of 2D hetero-structures for reducing the power consumption. The demonstrated FE-CCD provides a valuable strategy for versatile 2D materials to be monolithically integrated into the semiconductor imaging technology.

physics.app-ph↗
arXiv2017

The effect of heterogeneity on flocking behavior and systemic risk

The goal of this paper is to study organized flocking behavior and systemic risk in heterogeneous mean-field interacting diffusions. We illustrate in a number of case studies the effect of heterogeneity in the behavior of systemic risk in the system, i.e., the risk that several agents default simultaneously as a result of interconnections. We also investigate the effect of heterogeneity on the "flocking behavior" of different agents, i.e., when agents with different dynamics end up following very similar paths and follow closely the mean behavior of the system. Using Laplace asymptotics, we derive an asymptotic formula for the tail of the loss distribution as the number of agents grows to infinity. This characterizes the tail of the loss distribution and the effect of the heterogeneity of the network on the tail loss probability.

q-fin.RM↗
arXiv2015

Detecting Overlapping Communities from Local Spectral Subspaces

Based on the definition of local spectral subspace, we propose a novel approach called LOSP for local overlapping community detection. Instead of using the invariant subspace spanned by the dominant eigenvectors of the entire network, we run the power method for a few steps to approximate the leading eigenvectors that depict the embedding of the local neighborhood structure around seeds of interest. We then seek a sparse approximate indicator vector in the local spectral subspace spanned by these vectors such that the seeds are in its support. We evaluate LOSP on five large real world networks across various domains with labeled ground-truth communities and compare the results with the state-of-the-art community detection approaches. LOSP identifies the members of a target community with high accuracy from very few seed members, and outperforms the local Heat Kernel or PageRank diffusions as well as the global baselines. Two candidate definitions of the local spectral subspace are analyzed, and different community scoring functions for determining the community boundary, including two new metrics, are thoroughly evaluated. The structural properties of different seed sets and the impact of the seed set size are discussed. We observe low degree seeds behave better, and LOSP is robust even when started from a single random seed. Using LOSP as a subroutine and starting from each ego connected component, we try the harder yet significant task of identifying all communities a single vertex is in. Experiments show that the proposed method achieves high F1 measures on the detected multiple local overlapping communities containing the seed vertex.

cs.SI↗
arXiv2010

Software Requirements Specification of the IUfA's UUIS -- a Team 2 COMP5541-W10 Project Approach

In the 52-page document, we describe our approach to the Software Requirements Specification of the IUfA's UUIS prototype. This includes the overall system description, functional requirements, non-functional requirements, use cases, the corresponding data dictionary for all entities involved, mock user interface (UI) design, and the overall projected cost estimate. The design specification of UUIS can be found in arXiv:1005.0665.

cs.SE↗