arXiv subjects

Yitong Li

Publications and source records attributed to Yitong Li.

140 records · Page 8Linked to original sources

arXiv2019

Diffusion Maps for Textual Network Embedding

Textual network embedding leverages rich text information associated with the network to learn low-dimensional vectorial representations of vertices. Rather than using typical natural language processing (NLP) approaches, recent research exploits the relationship of texts on the same edge to graphically embed text. However, these models neglect to measure the complete level of connectivity between any two texts in the graph. We present diffusion maps for textual network embedding (DMTE), integrating global structural information of the graph to capture the semantic relatedness between texts, with a diffusion-convolution operation applied on the text inputs. In addition, a new objective function is designed to efficiently preserve the high-order proximity using the graph diffusion. Experimental results show that the proposed approach outperforms state-of-the-art methods on the vertex-classification and link-prediction tasks.

cs.CL↗

arXiv2019

Enhanced Welding Operator Quality Performance Measurement: Work Experience-Integrated Bayesian Prior Determination

Measurement of operator quality performance has been challenging in the construction fabrication industry. Among various causes, the learning effect is a significant factor, which needs to be incorporated in achieving a reliable operator quality performance analysis. This research aims to enhance a previously developed operator quality performance measurement approach by incorporating the learning effect (i.e., work experience). To achieve this goal, the Plateau learning model is selected to quantitatively represent the relationship between quality performance and work experience through a beta-binomial regression approach. Based on this relationship, an informative prior determination approach, which incorporates operator work experience information, is developed to enhance the previous Bayesian-based operator quality performance measurement. Academically, this research provides a systematic approach to derive Bayesian informative priors through integrating multi-source information. Practically, the proposed approach reliably measures operator quality performance in fabrication quality control processes.

stat.AP↗

arXiv2018

Learning Context-Sensitive Convolutional Filters for Text Processing

Convolutional neural networks (CNNs) have recently emerged as a popular building block for natural language processing (NLP). Despite their success, most existing CNN models employed in NLP share the same learned (and static) set of filters for all input sentences. In this paper, we consider an approach of using a small meta network to learn context-sensitive convolutional filters for text processing. The role of meta network is to abstract the contextual information of a sentence or document into a set of input-aware filters. We further generalize this framework to model sentence pairs, where a bidirectional filter generation mechanism is introduced to encapsulate co-dependent sentence representations. In our benchmarks on four different tasks, including ontology classification, sentiment analysis, answer sentence selection, and paraphrase identification, our proposed model, a modified CNN with context-sensitive filters, consistently outperforms the standard CNN and attention-based CNN baselines. By visualizing the learned context-sensitive filters, we further validate and rationalize the effectiveness of proposed framework.

cs.CL↗

arXiv2018

Multi-Label Learning from Medical Plain Text with Convolutional Residual Models

Predicting diagnoses from Electronic Health Records (EHRs) is an important medical application of multi-label learning. We propose a convolutional residual model for multi-label classification from doctor notes in EHR data. A given patient may have multiple diagnoses, and therefore multi-label learning is required. We employ a Convolutional Neural Network (CNN) to encode plain text into a fixed-length sentence embedding vector. Since diagnoses are typically correlated, a deep residual network is employed on top of the CNN encoder, to capture label (diagnosis) dependencies and incorporate information directly from the encoded sentence vector. A real EHR dataset is considered, and we compare the proposed model with several well-known baselines, to predict diagnoses based on doctor notes. Experimental results demonstrate the superiority of the proposed convolutional residual model.

stat.ML↗

arXiv2018

What's in a Domain? Learning Domain-Robust Text Representations using Adversarial Training

Most real world language problems require learning from heterogenous corpora, raising the problem of learning robust models which generalise well to both similar (in domain) and dissimilar (out of domain) instances to those seen in training. This requires learning an underlying task, while not learning irrelevant signals and biases specific to individual domains. We propose a novel method to optimise both in- and out-of-domain accuracy based on joint learning of a structured neural model with domain-specific and domain-general components, coupled with adversarial training for domain. Evaluating on multi-domain language identification and multi-domain sentiment analysis, we show substantial improvements over standard domain adaptation techniques, and domain-adversarial training.

cs.CL↗

arXiv2018

Towards Robust and Privacy-preserving Text Representations

Written text often provides sufficient clues to identify the author, their gender, age, and other important attributes. Consequently, the authorship of training and evaluation corpora can have unforeseen impacts, including differing model performance for different user groups, as well as privacy implications. In this paper, we propose an approach to explicitly obscure important author characteristics at training time, such that representations learned are invariant to these attributes. Evaluating on two tasks, we show that this leads to increased privacy in the learned representations, as well as more robust models to varying evaluation conditions, including out-of-domain corpora.

cs.CL↗

arXiv2018

Maximum Sum Rate of Slotted Aloha with Successive Interference Cancellation

This is a sequel of our previous work [8] on characterization of maximum sum rate of slotted Aloha networks. By extending the analysis to incorporate the capacity-achieving receiver structure, Successive Interference Cancellation (SIC), this paper aims to identify the rate loss due to random access. Specifically, two representative SIC receivers are considered, i.e, ordered SIC where packets are decoded in a descending order of their received power, and unordered SIC where packets are decoded in a random order. The maximum sum rate and the corresponding optimal parameter setting including the transmission probability and the information encoding rate in both cases are obtained as functions of the mean received signal-to-noise ratio (SNR). The comparison to the capture model shows that the gains are significant only with the ordered SIC at moderate values of the mean received SNR $ρ$. With a large $ρ$, the rate gap diminishes, and they all have the same high-SNR slope of $e^{-1}$, which is far below that of the ergodic sum capacity of fading channels. The effect of multipacket reception (MPR) on the sum rate performance is also studied by comparing the MPR receivers including SIC and the capture model to the classical collision model.

cs.IT↗

arXiv2017

Complexity Analysis Approach for Prefabricated Construction Products Using Uncertain Data Clustering

This paper proposes an uncertain data clustering approach to quantitatively analyze the complexity of prefabricated construction components through the integration of quality performance-based measures with associated engineering design information. The proposed model is constructed in three steps, which (1) measure prefabricated construction product complexity (hereafter referred to as product complexity) by introducing a Bayesian-based nonconforming quality performance indicator; (2) score each type of product complexity by developing a Hellinger distance-based distribution similarity measurement; and (3) cluster products into homogeneous complexity groups by using the agglomerative hierarchical clustering technique. An illustrative example is provided to demonstrate the proposed approach, and a case study of an industrial company in Edmonton, Canada, is conducted to validate the feasibility and applicability of the proposed model. This research inventively defines and investigates product complexity from the perspective of product quality performance with design information associated. The research outcomes provide simplified, interpretable, and informative insights for practitioners to better analyze and manage product complexity. In addition to this practical contribution, a novel hierarchical clustering technique is devised. This technique is capable of clustering uncertain data (i.e., beta distributions) with lower computational complexity and has the potential to be generalized to cluster all types of uncertain data.

cs.DB↗

arXiv2017

Video Generation From Text

Generating videos from text has proven to be a significant challenge for existing generative models. We tackle this problem by training a conditional generative model to extract both static and dynamic information from text. This is manifested in a hybrid framework, employing a Variational Autoencoder (VAE) and a Generative Adversarial Network (GAN). The static features, called "gist," are used to sketch text-conditioned background color and object layout structure. Dynamic features are considered by transforming input text into an image filter. To obtain a large amount of data for training the deep-learning model, we develop a method to automatically create a matched text-video corpus from publicly available online videos. Experimental results show that the proposed framework generates plausible and diverse videos, while accurately reflecting the input text information. It significantly outperforms baseline models that directly adapt text-to-image generation procedures to produce videos. Performance is evaluated both visually and by adapting the inception score used to evaluate image generation in GANs.

cs.MM↗

arXiv2016

Learning Robust Representations of Text

Deep neural networks have achieved remarkable results across many language processing tasks, however these methods are highly sensitive to noise and adversarial attacks. We present a regularization based method for limiting network sensitivity to its inputs, inspired by ideas from computer vision, thus learning models that are more robust. Empirical evaluation over a range of sentiment datasets with a convolutional neural network shows that, compared to a baseline model and the dropout method, our method achieves superior performance over noisy inputs and out-of-domain data.

cs.CL↗

arXiv2015

Maximum Sum Rate of Slotted Aloha with Capture

The sum rate performance of random-access networks crucially depends on the access protocol and receiver structure. Despite extensive studies, how to characterize the maximum sum rate of the simplest version of random access, Aloha, remains an open question. In this paper, a comprehensive study of the sum rate performance of slotted Aloha networks is presented. By extending the unified analytical framework proposed in [20], [21] from the classical collision model to the capture model, the network steady-state point in saturated conditions is derived as a function of the signal-to-interference-plus-noise ratio (SINR) threshold which determines a fundamental tradeoff between the information encoding rate and the network throughput. To maximize the sum rate, both the SINR threshold and backoff parameters of nodes should be properly selected. Explicit expressions of the maximum sum rate and the optimal setting are obtained, which show that similar to the sum capacity of the multiple access channel, the maximum sum rate of slotted Aloha also logarithmically increases with the mean received signal-to-noise ratio (SNR), but the high-SNR slope is only $e^{-1}$. Effects of backoff and power control on the sum rate performance of slotted Aloha networks are further discussed, which shed important light on the practical network design.

cs.IT↗

arXiv2015

A Survey of Heterogeneous Information Network Analysis

Most real systems consist of a large number of interacting, multi-typed components, while most contemporary researches model them as homogeneous networks, without distinguishing different types of objects and links in the networks. Recently, more and more researchers begin to consider these interconnected, multi-typed data as heterogeneous information networks, and develop structural analysis approaches by leveraging the rich semantic meaning of structural types of objects and links in the networks. Compared to widely studied homogeneous network, the heterogeneous information network contains richer structure and semantic information, which provides plenty of opportunities as well as a lot of challenges for data mining. In this paper, we provide a survey of heterogeneous information network analysis. We will introduce basic concepts of heterogeneous information network analysis, examine its developments on different data mining tasks, discuss some advanced topics, and point out some future research directions.

cs.SI↗

arXiv2015

EventNet: A Large Scale Structured Concept Library for Complex Event Detection in Video

Event-specific concepts are the semantic concepts designed for the events of interest, which can be used as a mid-level representation of complex events in videos. Existing methods only focus on defining event-specific concepts for a small number of predefined events, but cannot handle novel unseen events. This motivates us to build a large scale event-specific concept library that covers as many real-world events and their concepts as possible. Specifically, we choose WikiHow, an online forum containing a large number of how-to articles on human daily life events. We perform a coarse-to-fine event discovery process and discover 500 events from WikiHow articles. Then we use each event name as query to search YouTube and discover event-specific concepts from the tags of returned videos. After an automatic filter process, we end up with 95,321 videos and 4,490 concepts. We train a Convolutional Neural Network (CNN) model on the 95,321 videos over the 500 events, and use the model to extract deep learning feature from video content. With the learned deep learning feature, we train 4,490 binary SVM classifiers as the event-specific concept library. The concepts and events are further organized in a hierarchical structure defined by WikiHow, and the resultant concept library is called EventNet. Finally, the EventNet concept library is used to generate concept based representation of event videos. To the best of our knowledge, EventNet represents the first video event ontology that organizes events and their concepts into a semantic structure. It offers great potential for event retrieval and browsing. Extensive experiments over the zero-shot event retrieval task when no training samples are available show that the EventNet concept library consistently and significantly outperforms the state-of-the-art (such as the 20K ImageNet concepts trained with CNN) by a large margin up to 207%.

cs.CV↗

arXiv2014

HRank: A Path based Ranking Framework in Heterogeneous Information Network

Recently, there is a surge of interests on heterogeneous information network analysis. As a newly emerging network model, heterogeneous information networks have many unique features (e.g., complex structure and rich semantics) and a number of interesting data mining tasks have been exploited in this kind of networks, such as similarity measure, clustering, and classification. Although evaluating the importance of objects has been well studied in homogeneous networks, it is not yet exploited in heterogeneous networks. In this paper, we study the ranking problem in heterogeneous networks and propose the HRank framework to evaluate the importance of multiple types of objects and meta paths. Since the importance of objects depends upon the meta paths in heterogeneous networks, HRank develops a path based random walk process. Moreover, a constrained meta path is proposed to subtly capture the rich semantics in heterogeneous networks. Furthermore, HRank can simultaneously determine the importance of objects and meta paths through applying the tensor analysis. Extensive experiments on three real datasets show that HRank can effectively evaluate the importance of objects and paths together. Moreover, the constrained meta path shows its potential on mining subtle semantics by obtaining more accurate ranking results.

cs.IR↗