arXiv subjects

Xiao Yu

Publications and source records attributed to Xiao Yu.

96 records · Page 6Linked to original sources

arXiv2019

Adaptive Transfer Learning of Multi-View Time Series Classification

Time Series Classification (TSC) has been an important and challenging task in data mining, especially on multivariate time series and multi-view time series data sets. Meanwhile, transfer learning has been widely applied in computer vision and natural language processing applications to improve deep neural network's generalization capabilities. However, very few previous works applied transfer learning framework to time series mining problems. Particularly, the technique of measuring similarities between source domain and target domain based on dynamic representation such as density estimation with importance sampling has never been combined with transfer learning framework. In this paper, we first proposed a general adaptive transfer learning framework for multi-view time series data, which shows strong ability in storing inter-view importance value in the process of knowledge transfer. Next, we represented inter-view importance through some time series similarity measurements and approximated the posterior distribution in latent space for the importance sampling via density estimation techniques. We then computed the matrix norm of sampled importance value, which controls the degree of knowledge transfer in pre-training process. We further evaluated our work, applied it to many other time series classification tasks, and observed that our architecture maintained desirable generalization ability. Finally, we concluded that our framework could be adapted with deep learning techniques to receive significant model performance improvements.

cs.LG↗

arXiv2019

Adversarial Defense Framework for Graph Neural Network

Graph neural network (GNN), as a powerful representation learning model on graph data, attracts much attention across various disciplines. However, recent studies show that GNN is vulnerable to adversarial attacks. How to make GNN more robust? What are the key vulnerabilities in GNN? How to address the vulnerabilities and defense GNN against the adversarial attacks? In this paper, we propose DefNet, an effective adversarial defense framework for GNNs. In particular, we first investigate the latent vulnerabilities in every layer of GNNs and propose corresponding strategies including dual-stage aggregation and bottleneck perceptron. Then, to cope with the scarcity of training data, we propose an adversarial contrastive learning method to train the GNN in a conditional GAN manner by leveraging the high-level graph representation. Extensive experiments on three public datasets demonstrate the effectiveness of DefNet in improving the robustness of popular GNN variants, such as Graph Convolutional Network and GraphSAGE, under various types of adversarial attacks.

cs.LG↗

arXiv2018

Some estimates for the bilinear fractional integrals on the Morrey space

In this paper, we are interested in the following bilinear fractional integral operator $B\mathcal{I}_α$ defined by \[ B\mathcal{I}_α({f,g})(x)=\int_{% %TCIMACRO{\U{211d} }% %BeginExpansion \mathbb{R} %EndExpansion ^{n}}\frac{f(x-y)g(x+y)}{|y|^{n-α}}dy, \] with $0< α<n$. We prove the weighted boundedness of $B\mathcal{I}_α$ on the Morrey type spaces. Moreover, an Olsen type inequality for $B\mathcal{I}_α$ is also given.

math.CA↗

arXiv2017

Learning to Generate Posters of Scientific Papers by Probabilistic Graphical Models

Researchers often summarize their work in the form of scientific posters. Posters provide a coherent and efficient way to convey core ideas expressed in scientific papers. Generating a good scientific poster, however, is a complex and time consuming cognitive task, since such posters need to be readable, informative, and visually aesthetic. In this paper, for the first time, we study the challenging problem of learning to generate posters from scientific papers. To this end, a data-driven framework, that utilizes graphical models, is proposed. Specifically, given content to display, the key elements of a good poster, including attributes of each panel and arrangements of graphical elements are learned and inferred from data. During the inference stage, an MAP inference framework is employed to incorporate some design principles. In order to bridge the gap between panel attributes and the composition within each panel, we also propose a recursive page splitting algorithm to generate the panel layout for a poster. To learn and validate our model, we collect and release a new benchmark dataset, called NJU-Fudan Paper-Poster dataset, which consists of scientific papers and corresponding posters with exhaustively labelled panels and attributes. Qualitative and quantitative results indicate the effectiveness of our approach.

cs.CV↗

arXiv2015

Traveling waves and spreading speeds for time-space periodic monotone systems

The theory of traveling waves and spreading speeds is developed for time-space periodic monotone semiflows with monostable structure. By using traveling waves of the associated Poincaré maps in a strong sense, we establish the existence of time-space periodic traveling waves and spreading speeds. We then apply these abstract results to a two species competition reaction-advection-diffusion model. It turns out that the minimal wave speed exists and coincides with the single spreading speed for such a system no matter whether the spreading speed is linearly determinate. We also obtain a set of sufficient conditions for the spreading speed to be linearly determinate.

math.AP↗

arXiv2014

Propagation Phenomena for A Reaction-Advection-Diffusion Competition Model in A Periodic Habitat

This paper is devoted to the study of propagation phenomena for a Lotka-Volterra reaction-advection-diffusion competition model in a periodic habitat. We first investigate the global attractivity of a semi-trival steady state for the periodic initial value problem. Then we establish the existence of the rightward spreading speed and its coincidence with the minimal wave speed for spatially periodic rightward traveling waves. We also obtain a set of sufficient conditions for the rightward spreading speed to be linearly determinate. Finally, we apply the obtained results to a prototypical reaction-diffusion model.

math.AP↗