arXiv subjects

Ji Zhang

Publications and source records attributed to Ji Zhang.

231 records · Page 13Linked to original sources

arXiv2019

Method and Dataset Mining in Scientific Papers

Literature analysis facilitates researchers better understanding the development of science and technology. The conventional literature analysis focuses on the topics, authors, abstracts, keywords, references, etc., and rarely pays attention to the content of papers. In the field of machine learning, the involved methods (M) and datasets (D) are key information in papers. The extraction and mining of M and D are useful for discipline analysis and algorithm recommendation. In this paper, we propose a novel entity recognition model, called MDER, and constructe datasets from the papers of the PAKDD conferences (2009-2019). Some preliminary experiments are conducted to assess the extraction performance and the mining results are visualized.

cs.LG↗

arXiv2019

Following Social Groups: Socially Compliant Autonomous Navigation in Dense Crowds

In densely populated environments, socially compliant navigation is critical for autonomous robots as driving close to people is unavoidable. This manner of social navigation is challenging given the constraints of human comfort and social rules. Traditional methods based on hand-craft cost functions to achieve this task have difficulties to operate in the complex real world. Other learning-based approaches fail to address the naturalness aspect from the perspective of collective formation behaviors. We present an autonomous navigation system capable of operating in dense crowds and utilizing information of social groups. The underlying system incorporates a deep neural network to track social groups and join the flow of a social group in facilitating the navigation. A collision avoidance layer in the system further ensures navigation safety. In experiments, our method generates socially compliant behaviors as state-of-the-art methods. More importantly, the system is capable of navigating safely in a densely populated area (10+ people in a 10m x 20m area) following crowd flows to reach the goal.

cs.RO↗

arXiv2019

Large-Scale Visual Relationship Understanding

Large scale visual understanding is challenging, as it requires a model to handle the widely-spread and imbalanced distribution of triples. In real-world scenarios with large numbers of objects and relations, some are seen very commonly while others are barely seen. We develop a new relationship detection model that embeds objects and relations into two vector spaces where both discriminative capability and semantic affinity are preserved. We learn both a visual and a semantic module that map features from the two modalities into a shared space, where matched pairs of features have to discriminate against those unmatched, but also maintain close distances to semantically similar ones. Benefiting from that, our model can achieve superior performance even when the visual entity categories scale up to more than 80,000, with extremely skewed class distribution. We demonstrate the efficacy of our model on a large and imbalanced benchmark based of Visual Genome that comprises 53,000+ objects and 29,000+ relations, a scale at which no previous work has ever been evaluated at. We show superiority of our model over carefully designed baselines on the original Visual Genome dataset with 80,000+ categories. We also show state-of-the-art performance on the VRD dataset and the scene graph dataset which is a subset of Visual Genome with 200 categories.

cs.CV↗

arXiv2019

Graphical Contrastive Losses for Scene Graph Parsing

Most scene graph parsers use a two-stage pipeline to detect visual relationships: the first stage detects entities, and the second predicts the predicate for each entity pair using a softmax distribution. We find that such pipelines, trained with only a cross entropy loss over predicate classes, suffer from two common errors. The first, Entity Instance Confusion, occurs when the model confuses multiple instances of the same type of entity (e.g. multiple cups). The second, Proximal Relationship Ambiguity, arises when multiple subject-predicate-object triplets appear in close proximity with the same predicate, and the model struggles to infer the correct subject-object pairings (e.g. mis-pairing musicians and their instruments). We propose a set of contrastive loss formulations that specifically target these types of errors within the scene graph parsing problem, collectively termed the Graphical Contrastive Losses. These losses explicitly force the model to disambiguate related and unrelated instances through margin constraints specific to each type of confusion. We further construct a relationship detector, called RelDN, using the aforementioned pipeline to demonstrate the efficacy of our proposed losses. Our model outperforms the winning method of the OpenImages Relationship Detection Challenge by 4.7\% (16.5\% relative) on the test set. We also show improved results over the best previous methods on the Visual Genome and Visual Relationship Detection datasets.

cs.CV↗

arXiv2019

2nd Place Solution to the GQA Challenge 2019

We present a simple method that achieves unexpectedly superior performance for Complex Reasoning involved Visual Question Answering. Our solution collects statistical features from high-frequency words of all the questions asked about an image and use them as accurate knowledge for answering further questions of the same image. We are fully aware that this setting is not ubiquitously applicable, and in a more common setting one should assume the questions are asked separately and they cannot be gathered to obtain a knowledge base. Nonetheless, we use this method as an evidence to demonstrate our observation that the bottleneck effect is more severe on the feature extraction part than it is on the knowledge reasoning part. We show significant gaps when using the same reasoning model with 1) ground-truth features; 2) statistical features; 3) detected features from completely learned detectors, and analyze what these gaps mean to researches on visual reasoning topics. Our model with the statistical features achieves the 2nd place in the GQA Challenge 2019.

cs.CV↗

arXiv2019

Impact of Electrostatic Doping Level on the Dissipative Transport in Graphene Nanoribbons Tunnel Field-Effect Transistors

The impact of electrostatic doping level on the dissipative transport of Armchair GNR-TFET is studied using the Quantum Perturbation Theory (QPT) with the Extended Lowest Order Expansion (XLOE) implementation method. Results show that the doping level of the source and drain sides of the GNR-TFET has a significant impact on the phonon contribution to the carrier transport process. Unlike in other similar studies, where phonons are believed to have a constant detrimental influence on the ION/IOFF ratio and Subthreshold Swing (SS) of the TFET devices due to the phonon absorption-assisted tunneling, we show that by a proper engineering of the doping level in the source and drain, the phonon absorption assisted tunneling can be effectively inhibited. We also show that as temperature increase, the device switching property deteriorates in both the ballistic and dissipative transport regimes, and there exists a temperature-dependent critical doping level where the device has optimal switching behavior.

cond-mat.mes-hall↗

arXiv2019

LCuts: Linear Clustering of Bacteria using Recursive Graph Cuts

Bacterial biofilm segmentation poses significant challenges due to lack of apparent structure, poor imaging resolution, limited contrast between conterminous cells and high density of cells that overlap. Although there exist bacterial segmentation algorithms in the existing art, they fail to delineate cells in dense biofilms, especially in 3D imaging scenarios in which the cells are growing and subdividing in a complex manner. A graph-based data clustering method, LCuts, is presented with the application on bacterial cell segmentation. By constructing a weighted graph with node features in locations and principal orientations, the proposed method can automatically classify and detect differently oriented aggregations of linear structures (represent by bacteria in the application). The method assists in the assessment of several facets, such as bacterium tracking, cluster growth, and mapping of migration patterns of bacterial biofilms. Quantitative and qualitative measures for 2D data demonstrate the superiority of proposed method over the state of the art. Preliminary 3D results exhibit reliable classification of the cells with 97% accuracy.

eess.IV↗

arXiv2019

Efficient and Secure ECDSA Algorithm and its Applications: A Survey

Public-key cryptography algorithms, especially elliptic curve cryptography (ECC) and elliptic curve digital signature algorithm (ECDSA) have been attracting attention from many researchers in different institutions because these algorithms provide security and high performance when being used in many areas such as electronic-healthcare, electronic-banking, electronic-commerce, electronic-vehicular, and electronic-governance. These algorithms heighten security against various attacks and at the same time improve performance to obtain efficiencies (time, memory, reduced computation complexity, and energy saving) in an environment of the constrained source and large systems. This paper presents detailed and a comprehensive survey of an update of the ECDSA algorithm in terms of performance, security, and applications.

cs.CR↗

arXiv2019

RAMHU: A New Robust Lightweight Scheme for Mutual Users Authentication in Healthcare Applications

Providing a mechanism to authenticate users in healthcare applications is an essential security requirement to prevent both external and internal attackers from penetrating patients' identities and revealing their health data. Many schemes have been developed to provide authentication mechanisms to ensure that only legitimate users are authorized to connect, but these schemes still suffer from vulnerable security. Various attacks expose patients' data for malicious tampering or destruction. Transferring health-related data and information between users and the health centre makes them exposed to penetration by adversaries as they may move through an insecure channel. In addition, previous mechanisms have suffered from the poor protection of users' authentication information. To ensure the protection of patients' information and data, we propose a scheme that authenticates users based on the information of both the device and the legitimate user. In this paper, we propose a Robust Authentication Model for Healthcare Users (RAMHU) that provides mutual authentication between server and clients. This model utilizes an Elliptic Curve Integrated Encryption Scheme (ECIES) and PHOTON to achieve strong security and a good overall performance. RAMHU relies on multi pseudonyms, physical address, and one-time password mechanisms to authenticate legitimate users. Moreover, extensive informal and formal security analysis with the automated validation of Internet security protocols and applications (AVISPA) tool demonstrates that our model offers a high level of security in repelling a wide variety of possible attacks.

cs.CR↗

arXiv2018

A Deep Cascade Model for Multi-Document Reading Comprehension

A fundamental trade-off between effectiveness and efficiency needs to be balanced when designing an online question answering system. Effectiveness comes from sophisticated functions such as extractive machine reading comprehension (MRC), while efficiency is obtained from improvements in preliminary retrieval components such as candidate document selection and paragraph ranking. Given the complexity of the real-world multi-document MRC scenario, it is difficult to jointly optimize both in an end-to-end system. To address this problem, we develop a novel deep cascade learning model, which progressively evolves from the document-level and paragraph-level ranking of candidate texts to more precise answer extraction with machine reading comprehension. Specifically, irrelevant documents and paragraphs are first filtered out with simple functions for efficiency consideration. Then we jointly train three modules on the remaining texts for better tracking the answer: the document extraction, the paragraph extraction and the answer extraction. Experiment results show that the proposed method outperforms the previous state-of-the-art methods on two large-scale multi-document benchmark datasets, i.e., TriviaQA and DuReader. In addition, our online system can stably serve typical scenarios with millions of daily requests in less than 50ms.

cs.CL↗

arXiv2018

An Interpretable Model for Scene Graph Generation

We propose an efficient and interpretable scene graph generator. We consider three types of features: visual, spatial and semantic, and we use a late fusion strategy such that each feature's contribution can be explicitly investigated. We study the key factors about these features that have the most impact on the performance, and also visualize the learned visual features for relationships and investigate the efficacy of our model. We won the champion of the OpenImages Visual Relationship Detection Challenge on Kaggle, where we outperform the 2nd place by 5\% (20\% relatively). We believe an accurate scene graph generator is a fundamental stepping stone for higher-level vision-language tasks such as image captioning and visual QA, since it provides a semantic, structured comprehension of an image that is beyond pixels and objects.

cs.CV↗

arXiv2018

Introduction to the 1st Place Winning Model of OpenImages Relationship Detection Challenge

This article describes the model we built that achieved 1st place in the OpenImage Visual Relationship Detection Challenge on Kaggle. Three key factors contribute the most to our success: 1) language bias is a powerful baseline for this task. We build the empirical distribution $P(predicate|subject,object)$ in the training set and directly use that in testing. This baseline achieved the 2nd place when submitted; 2) spatial features are as important as visual features, especially for spatial relationships such as "under" and "inside of"; 3) It is a very effective way to fuse different features by first building separate modules for each of them, then adding their output logits before the final softmax layer. We show in ablation study that each factor can improve the performance to a non-trivial extent, and the model reaches optimal when all of them are combined.

cs.CV↗

arXiv2018

Improving Multilingual Semantic Textual Similarity with Shared Sentence Encoder for Low-resource Languages

Measuring the semantic similarity between two sentences (or Semantic Textual Similarity - STS) is fundamental in many NLP applications. Despite the remarkable results in supervised settings with adequate labeling, little attention has been paid to this task in low-resource languages with insufficient labeling. Existing approaches mostly leverage machine translation techniques to translate sentences into rich-resource language. These approaches either beget language biases, or be impractical in industrial applications where spoken language scenario is more often and rigorous efficiency is required. In this work, we propose a multilingual framework to tackle the STS task in a low-resource language e.g. Spanish, Arabic , Indonesian and Thai, by utilizing the rich annotation data in a rich resource language, e.g. English. Our approach is extended from a basic monolingual STS framework to a shared multilingual encoder pretrained with translation task to incorporate rich-resource language data. By exploiting the nature of a shared multilingual encoder, one sentence can have multiple representations for different target translation language, which are used in an ensemble model to improve similarity evaluation. We demonstrate the superiority of our method over other state of the art approaches on SemEval STS task by its significant improvement on non-MT method, as well as an online industrial product where MT method fails to beat baseline while our approach still has consistently improvements.

cs.CL↗

arXiv2018

Semi-Autoregressive Neural Machine Translation

Existing approaches to neural machine translation are typically autoregressive models. While these models attain state-of-the-art translation quality, they are suffering from low parallelizability and thus slow at decoding long sequences. In this paper, we propose a novel model for fast sequence generation --- the semi-autoregressive Transformer (SAT). The SAT keeps the autoregressive property in global but relieves in local and thus is able to produce multiple successive words in parallel at each time step. Experiments conducted on English-German and Chinese-English translation tasks show that the SAT achieves a good balance between translation quality and decoding speed. On WMT'14 English-German translation, the SAT achieves 5.58$\times$ speedup while maintains 88\% translation quality, significantly better than the previous non-autoregressive methods. When produces two words at each time step, the SAT is almost lossless (only 1\% degeneration in BLEU score).

cs.CL↗

arXiv2018

Exploiting Effective Representations for Chinese Sentiment Analysis Using a Multi-Channel Convolutional Neural Network

Effective representation of a text is critical for various natural language processing tasks. For the particular task of Chinese sentiment analysis, it is important to understand and choose an effective representation of a text from different forms of Chinese representations such as word, character and pinyin. This paper presents a systematic study of the effect of these representations for Chinese sentiment analysis by proposing a multi-channel convolutional neural network (MCCNN), where each channel corresponds to a representation. Experimental results show that: (1) Word wins on the dataset of low OOV rate while character wins otherwise; (2) Using these representations in combination generally improves the performance; (3) The representations based on MCCNN outperform conventional ngram features using SVM; (4) The proposed MCCNN model achieves the competitive performance against the state-of-the-art model fastText for Chinese sentiment analysis.

cs.CL↗