Search

Your search keyword '"Lu, Shuqiang"' showing total 38 results
Start Over Author "Lu, Shuqiang"
38 results on '"Lu, Shuqiang"'

1. Evaluating and Improving Adversarial Robustness of Machine Learning-Based Network Intrusion Detectors

Author

Han, Dongqi, Wang, Zhiliang, Zhong, Ying, Chen, Wenqi, Yang, Jiahai, Lu, Shuqiang, Shi, Xingang, and Yin, Xia

Subjects
Computer Science - Cryptography and Security, Computer Science - Machine Learning, Computer Science - Networking and Internet Architecture
Abstract

Machine learning (ML), especially deep learning (DL) techniques have been increasingly used in anomaly-based network intrusion detection systems (NIDS). However, ML/DL has shown to be extremely vulnerable to adversarial attacks, especially in such security-sensitive systems. Many adversarial attacks have been proposed to evaluate the robustness of ML-based NIDSs. Unfortunately, existing attacks mostly focused on feature-space and/or white-box attacks, which make impractical assumptions in real-world scenarios, leaving the study on practical gray/black-box attacks largely unexplored. To bridge this gap, we conduct the first systematic study of the gray/black-box traffic-space adversarial attacks to evaluate the robustness of ML-based NIDSs. Our work outperforms previous ones in the following aspects: (i) practical-the proposed attack can automatically mutate original traffic with extremely limited knowledge and affordable overhead while preserving its functionality; (ii) generic-the proposed attack is effective for evaluating the robustness of various NIDSs using diverse ML/DL models and non-payload-based features; (iii) explainable-we propose an explanation method for the fragile robustness of ML-based NIDSs. Based on this, we also propose a defense scheme against adversarial attacks to improve system robustness. We extensively evaluate the robustness of various NIDSs using diverse feature sets and ML/DL models. Experimental results show our attack is effective (e.g., >97% evasion rate in half cases for Kitsune, a state-of-the-art NIDS) with affordable execution cost and the proposed defense method can effectively mitigate such attacks (evasion rate is reduced by >50% in most cases)., Comment: This article has been accepted for publication by IEEE JSAC

Published
2020
Full Text
View/download PDF

2. New Era of Deeplearning-Based Malware Intrusion Detection: The Malware Detection and Prediction Based On Deep Learning

Author

Lu, Shuqiang, Ying, Lingyun, Lin, Wenjie, Wang, Yu, Nie, Meining, Shen, Kaiwen, Liu, Lu, and Duan, Haixin

Subjects
Computer Science - Cryptography and Security, Computer Science - Machine Learning
Abstract

With the development of artificial intelligence algorithms like deep learning models and the successful applications in many different fields, further similar trails of deep learning technology have been made in cyber security area. It shows the preferable performance not only in academic security research but also in industry practices when dealing with part of cyber security issues by deep learning methods compared to those conventional rules. Especially for the malware detection and classification tasks, it saves generous time cost and promotes the accuracy for a total pipeline of malware detection system. In this paper, we construct special deep neural network, ie, MalDeepNet (TB-Malnet and IB-Malnet) for malware dynamic behavior classification tasks. Then we build the family clustering algorithm based on deep learning and fulfil related testing. Except that, we also design a novel malware prediction model which could detect the malware coming in future through the Mal Generative Adversarial Network (Mal-GAN) implementation. All those algorithms present fairly considerable value in related datasets afterwards., Comment: Unable to open due to ethical problem

Published
2019

25. Development of a Deep Learning–Based Model for Diagnosing Breast Nodules With Ultrasound.

Author

Li, Jianming, Bu, Yunyun, Lu, Shuqiang, Pang, Hao, Luo, Chang, Liu, Yujiang, and Qian, Linxue

Subjects
BREAST ultrasound, RECEIVER operating characteristic curves, EXPERT systems, ARTIFICIAL intelligence, MAGNETIC resonance mammography, DEEP learning, BREAST
Abstract

Objectives: Artificial intelligence (AI) has been an important addition to medicine. We aimed to explore the use of deep learning (DL) to distinguish benign from malignant lesions with breast ultrasound (BUS). Methods: The DL model was trained with BUS nodule data using a standard protocol (1271 malignant nodules, 1053 benign nodules, and 2144 images of the contralateral normal breast). The model was tested with 692 images of 256 breast nodules. We used the accuracy, precision, recall, harmonic mean of recall and precision, and mean average precision as the indices to assess the DL model. We used 100 BUS images to evaluate differences in diagnostic accuracy among the AI system, experts (>25 years of experience), and physicians with varying levels of experience. A receiver operating characteristic curve was generated to evaluate the accuracy for distinguishing between benign and malignant breast nodules. Results: The DL model showed 73.3% sensitivity and 94.9% specificity for the diagnosis of benign versus malignant breast nodules (area under the curve, 0.943). No significant difference in diagnostic ability was found between the AI system and the expert group (P =.951), although the physicians with lower levels of experience showed significant differences from the AI and expert groups (P =.01 and.03, respectively). Conclusions: Deep learning could distinguish between benign and malignant breast nodules with BUS. On BUS images, DL achieved diagnostic accuracy equivalent to that of expert physicians. [ABSTRACT FROM AUTHOR]

Published
2021
Full Text
View/download PDF

26. Use of Monitoring Data to Interpret Active Landslide Movements and Hydrological Triggers in Three Gorges Reservoir

Author

Haifeng Huang, Guodong Zhang, Wu Yi, Yi Qinglin, and Lu Shuqiang

Subjects
010504 meteorology & atmospheric sciences, business.industry, Movement (music), 0211 other engineering and technologies, Landslide, 02 engineering and technology, Building and Construction, Geodesy, 01 natural sciences, Monitoring data, Temporal resolution, Global Positioning System, Geotechnical engineering, Safety, Risk, Reliability and Quality, business, Geology, 021101 geological & geomatics engineering, 0105 earth and related environmental sciences, Civil and Structural Engineering, Three gorges
Abstract

The primary objective of this research is to study the relationship between surface movement and hydrological triggers of a large and active landslide in the Three Gorges Reservoir (the Shuping landslide). When the landslide was reactivated in June 2003 owing to the impoundment of the Three Gorges Dam, measurement of surface movement began through the use of manual global positioning system (GPS) monitoring. Since June 2012, near-real-time data with fine temporal resolution have been collected through the use of continuous automatic GPS. In total, more than 10 years of surface movement have been recorded for the Shuping landslide. Patterns of surface movement have been spatially identified first, and then subsequently over time. A fast moving zone can be spatially identified from the primary deformation zone. The temporal evolution of the landslide consists of a typical stepwise progression in time with short periods of fast movement (FM) and longer periods of slower movement. Eight of the former ...

Published
2016

27. Properties of the System Operator of a Two-Unit Standby Redundant System

Author

Xing Qiao, Dan Ma, Yuanxia Sha, Lu Shuqiang, and Hui Xu

Subjects
Mathematical optimization, Operator (computer programming), Multiplication operator, Weak operator topology, Cofinal, Distribution (number theory), Spectrum (functional analysis), Topology, Domain (mathematical analysis), Mathematics, Resolvent
Abstract

In this paper, we investigate a two-unit standby redundant electronic equipment system under human failure. Using functional analysis and C0-semigroup theory, we first prove the system operator is a densely defined resolvent positive operator. Then, we obtain the ad joint operator of the system operator and its domain and the spectra distribution of the system operator. Furthermore, we prove that 0 is the growth bound of the system operator. Finally, we show that 0 is also the upper spectral bound of the system operator using the concept of co final and relative theory.

Published
2015

28. Real-Time Monitoring and Data Analysis of an Active Landslide in the Three Gorges Reservoir

Author

Wu Yi, Yi Qinglin, Lu Shuqiang, and Haifeng Huang

Subjects
Hydrology, Geography, Mining engineering, business.industry, Reservoir water, Monitoring data, Geological disaster, Global Positioning System, Monitoring system, Landslide, business, Three gorges
Abstract

Global Positioning System (GPS) real-time monitoring is beginning to replace early manual monitoring of landslides in the Three Gorges reservoir area, China. This paper takes a typical active reservoir-type slide as an example. Firstly, it introduces the real-time GPS monitoring system which includes field continuous GPS sensors, communication infrastructure and control center. Then based on the real-time monitoring data for a year, the temporal and spatial moving patterns are defined. Furthermore, by analyzing the relationship between landslide movement and reservoir water fluctuation, the most dangerous hydrologic condition to trigger the landslide deformation was found. The results indicate that real-time monitoring can greatly improve the precision of understanding landslide behavior and motion law, which is vital for the prevention and mitigation of geological disasters.

Published
2015

32. The real time image merge method for the remote sensing image acquired from the UAV

Author

Gao Pengqi, Lu Shuqiang, Zhao Hongying, and Yan Lei

Subjects
Image fusion, Geographic information system, Pixel, business.industry, Computer science, Gauss, Object detection, symbols.namesake, Pyramid, symbols, Computer vision, Artificial intelligence, business, Merge (version control), Gaussian process, Image resolution, Remote sensing, Feature detection (computer vision)
Abstract

One method is presented for moisacing the remote sensing image in this paper from the UAV.The method includes two parts:the Gauss Pyramid and the gray projection .This method should reduce the calculation amount and promote the speed. The experimentation result shows the method is effective for the UAV images.

Published
2005

37. Fe40Co40Se20Glassy Films Supported on Carbon Fiber Paper as Electrocatalysts in the Oxygen Evolution Reaction

Author

Lu, Shuqiang, Lian, Jun, Zhang, Fabao, Jiang, Wei, Hu, Qingzhuo, Li, Dongdong, and Zhang, Bo

Abstract

We report a type of transition metal selenide (Fe40Co40Se20) thin films supported on carbon fiber paper (CFP) (Fe40Co40Se20/CFP) and study the effects of the substrate temperature during thin film deposition on the electrocatalytic performance of the films in the oxygen evolution reaction. Fe40Co40Se20/CFP thin films were deposited by vacuum magnetron sputtering at substrate temperatures of 308, 393, and 573 K. In 1 M KOH, the amorphous film deposited at 393 K exhibited better comprehensive electrocatalytic activity and stability than that of the Fe40Co40Se20/CFP thin films deposited under other conditions, achieving a low overpotential of 307 mV at a current density of 10 mA cm−2and low Tafel slope of 35 mV dec−1. These results reveal that an appropriate substrate temperature during film formation improves the catalytic performance of thin film catalysts to some extent. A combination of low charge resistance and large electrochemically active surface area contributes to the electrocatalytic performance of the Fe40Co40Se20/CFP thin films.

Published
2019
Full Text
View/download PDF

38. [Effect of selective gut decontamination in regulation of inflammatory reaction in patients with systemic inflammatory response syndrome].

Author

Wang Y, Lu S, Gao J, Wang Z, and Cai A

Subjects
Decontamination, Humans, Interferon-gamma, Interleukin-1, Interleukin-8, Lipopolysaccharides, Prospective Studies, Tumor Necrosis Factor-alpha, Systemic Inflammatory Response Syndrome
Abstract

Objective: To explore the effect of selective gut decontamination in regulation of inflammatory reaction compared with rhubarb and glycerine enema for catharsis in patients with systemic inflammatory response syndrome ( SIRS ), and to discuss its mechanisms., Methods: A prospective randomized controlled trial was conducted. Fifty-seven patients with SIRS admitted to Department of General Surgery of Aviation General Hospital from June 2009 to June 2014 were enrolled. The patients were randomly divided into rhubarb decontaminate group, traditional decontaminate group and blank control group, with 19 cases in each group. Besides the treatment for primary disease, including anti-infection, operation, alleviate pain, nutritional support, and maintaining water and electrolyte balance, the patients in rhubarb decontaminate group received aqueous extract from rhubarb 15-20 g by gastric tube, enema, or peros, twice a day; and those in traditional decontaminate group received glycerine enema or glycerol enema, twice a day; while no gavage or enema was prescribed in blank control group. Peripheral blood was collected before and 72 hours after treatment. Enzyme linked immunosorbent assay (ELISA) was used to determine the concentration of lipopolysaccharide (LPS) and inflammatory mediators., Results: Compared with blank control group and traditional decontaminate group, the levels of interleukins (IL-1, IL-8), LPS, platelet activating factor (PAF), tumor necrosis factor-α (TNF-α), and γ- interferon (IFN-γ) before treatment was similar to that of rhubarb decontaminate group [IL-1 (ng/L): 53.154±5.783, 50.564±5.771, 51.082±6.403, F=0.994, P=0.377; IL-8 (ng/L): 70.492±6.146, 68.376±6.112, 68.673±8.384, F=0.514, P=0.601; LPS (μg/L): 11.630±2.449, 10.858±2.307, 10.463±2.145, F=1.261, P=0.291; PAF (μg/L): 4.173±0.395, 4.051±0.362, 4.078±0.487, F=0.446, P=0.642; TNF-α (ng/L): 132.498±10.772, 129.735±12.881, 127.207±11.514, F=0.963, P=0.388; IFN-γ (μg/L): 45.645±4.558, 43.692±5.578, 43.767±5.028, F=0.904, P=0.411]. The above parameters after treatment were significantly lower than those before treatment in three groups. The effect on the LPS and pro-inflammatory factors of the rhubarb decontaminate group was more obvious than that of the blank control group and traditional decontaminate group [LPS(μg/L): 7.571±1.113 vs. 9.008±1.904, 8.874±1.808, F=4.416, P=0.017; IL-1 (ng/L): 45.309±3.563 vs. 48.731±4.466, 46.112±4.322, F=3.557, P=0.035; IL-8 (ng/L): 60.492±5.346 vs. 65.553±5.384, 63.437±5.462, F=4.213, P=0.020; PAF (μg/L): 3.519±0.250 vs. 3.832±0.356, 3.766±0.309, F=5.450, P=0.007; TNF-α (ng/L): 114.988±8.772 vs. 123.230±10.433, 118.534±9.519, F=3.525, P=0.036; IFN-γ (μg/L): 38.683±3.190 vs. 41.831±4.122, 39.161±3.972, F=3.820, P=0.028]., Conclusions: The usage of selective gut decontamination can inhibit the release of endotoxin and inflammatory mediators in patients with SIRS, and it will get a better effect using rhubarb, and the mechanism may be related to the protection of intestinal mucosal barrier function.

Published
2015
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results