Filters








4,129 Hits in 2.0 sec

Project Evaluation Method Based on Matter-Element and Hierarchy Model

Haifeng Li Haifeng Li
2012 TELKOMNIKA (Telecommunication Computing Electronics and Control)  
In this paper, we define a ISSN: 1693-6930 Project Evaluation Method based on Matter-Element and Hierarchy Model (Haifeng Li) 587 comprehensive evaluation index system for the project to be reviewed.  ...  Method based on Matter-Element and Hierarchy Model (Haifeng Li) the project to be reviewed; C is the whole characteristics of the comprehensive index b k (k=1,2,…,m) of N o , namely, all the single indexes  ... 
doi:10.12928/telkomnika.v10i3.841 fatcat:zzwvnecrmradvps65qwytgrwxq

Understanding the Importance of Single Directions via Representative Substitution [article]

Li Chen, Hailun Ding, Qi Li, Zhuo Li, Jian Peng, Haifeng Li
2019 arXiv   pre-print
Understanding the internal representations of deep neural networks (DNNs) is crucal to explain their behavior. The interpretation of individual units, which are neurons in MLPs or convolution kernels in convolutional networks, has been paid much attention given their fundamental role. However, recent research (Morcos et al. 2018) presented a counterintuitive phenomenon, which suggests that an individual unit with high class selectivity, called interpretable units, has poor contributions to
more » ... alization of DNNs. In this work, we provide a new perspective to understand this counterintuitive phenomenon, which makes sense when we introduce Representative Substitution (RS). Instead of individually selective units with classes, the RS refers to the independence of a unit's representations in the same layer without any annotation. Our experiments demonstrate that interpretable units have high RS which are not critical to network's generalization. The RS provides new insights into the interpretation of DNNs and suggests that we need to focus on the independence and relationship of the representations.
arXiv:1911.05586v1 fatcat:soqnrbzwqrgqxbd4p6wq3mr5my

Liquid Crystalline Polymers [chapter]

Xiao Li, Haifeng Yu
2015 Plastics Engineering  
Another incentive for developing LCP films lies in the increasingly attractive barrier properties that are being uncovered.  ...  The primary difficulty in developing a comprehensive picture of the thermodynamic aspects of LCPs lies in the fact that a high degree of coupling exists between all degrees of freedom in the system, internal  ... 
doi:10.1201/b19190-12 fatcat:skytcrcli5h6tlibdbmj2nkxq4

AutoOD: Automated Outlier Detection via Curiosity-guided Search and Self-imitation Learning [article]

Yuening Li, Zhengzhang Chen, Daochen Zha, Kaixiong Zhou, Haifeng Jin, Haifeng Chen, Xia Hu
2020 arXiv   pre-print
Outlier detection is an important data mining task with numerous practical applications such as intrusion detection, credit card fraud detection, and video surveillance. However, given a specific complicated task with big data, the process of building a powerful deep learning based system for outlier detection still highly relies on human expertise and laboring trials. Although Neural Architecture Search (NAS) has shown its promise in discovering effective deep architectures in various domains,
more » ... such as image classification, object detection, and semantic segmentation, contemporary NAS methods are not suitable for outlier detection due to the lack of intrinsic search space, unstable search process, and low sample efficiency. To bridge the gap, in this paper, we propose AutoOD, an automated outlier detection framework, which aims to search for an optimal neural network model within a predefined search space. Specifically, we firstly design a curiosity-guided search strategy to overcome the curse of local optimality. A controller, which acts as a search agent, is encouraged to take actions to maximize the information gain about the controller's internal belief. We further introduce an experience replay mechanism based on self-imitation learning to improve the sample efficiency. Experimental results on various real-world benchmark datasets demonstrate that the deep model identified by AutoOD achieves the best performance, comparing with existing handcrafted models and traditional search methods.
arXiv:2006.11321v1 fatcat:sqptlaxro5gb7frxdugtvje46y

Adversarial Example in Remote Sensing Image Recognition [article]

Li Chen, Guowei Zhu, Qi Li, Haifeng Li
2020 arXiv   pre-print
Chen, Guowei Zhu and Haifeng Li are with the School of Geosciences and Info-Physics, Central South University, South Lushan Road, Changsha, 410083, China.  ...  Haifeng Li received the master's degree in transportation engineering from the South China University of Technology, Guangzhou, China, in 2005, and the Ph.D. degree in photogrammetry and remote sensing  ... 
arXiv:1910.13222v2 fatcat:ja4bgf6xazaq7klj7ct5ojsm5a

Synthesis of Europium-Doped Fluorapatite Nanorods and Their Biomedical Applications in Drug Delivery

Haifeng Zeng, Xiyu Li, Muyang Sun, Sufan Wu, Haifeng Chen
2017 Molecules  
Europium (Eu)-doped fluorapatite (FA) nanorods have a biocompatibility similar to that of hydroxyapatite (HA) for use as cell imaging biomaterials due to their luminescent property. Here, we discuss the new application of europium-doped fluorapatite (Eu-FA) nanorods as an anticancer drug carrier. The Eu-FA nanorods were prepared by using a hydrothermal method. The morphology, crystal structure, fluorescence, and composition were investigated. The specific crystal structure enables the effective
more » ... loading of drug molecules. Doxorubicin (DOX), which was used as a model anticancer drug, effectively loaded onto the surface of the nanorods. The DOX release was pH-dependent and occurred more rapidly at pH 5.5 than at pH 7.4. The intracellular penetration of the DOX-loaded Eu-FA nanorods (Eu-FA/DOX) can be imaged in situ due to the self-fluorescence property. Treatment of melanoma A375 cells with Eu-FA/DOX elicited a more effective apoptosis rate than direct DOX treatment. Overall, Eu-FA exhibits potential for tracking and treating tumors and may be potentially useful as a multifunctional carrier system to effectively load and sustainably deliver drugs.
doi:10.3390/molecules22050753 pmid:28481233 fatcat:3qfjas3zpjdy5j4zdrnwfbrw2u

Convolution Neural Network Architecture Learning for Remote Sensing Scene Classification [article]

Jie Chen, Haozhe Huang, Jian Peng, Jiawei Zhu, Li Chen, Wenbo Li, Binyu Sun, Haifeng Li
2020 arXiv   pre-print
Remote sensing image scene classification is a fundamental but challenging task in understanding remote sensing images. Recently, deep learning-based methods, especially convolutional neural network-based (CNN-based) methods have shown enormous potential to understand remote sensing images. CNN-based methods meet with success by utilizing features learned from data rather than features designed manually. The feature-learning procedure of CNN largely depends on the architecture of CNN. However,
more » ... ost of the architectures of CNN used for remote sensing scene classification are still designed by hand which demands a considerable amount of architecture engineering skills and domain knowledge, and it may not play CNN's maximum potential on a special dataset. In this paper, we proposed an automatically architecture learning procedure for remote sensing scene classification. We designed a parameters space in which every set of parameters represents a certain architecture of CNN (i.e., some parameters represent the type of operators used in the architecture such as convolution, pooling, no connection or identity, and the others represent the way how these operators connect). To discover the optimal set of parameters for a given dataset, we introduced a learning strategy which can allow efficient search in the architecture space by means of gradient descent. An architecture generator finally maps the set of parameters into the CNN used in our experiments.
arXiv:2001.09614v1 fatcat:drfzcrelbrhgngujdgrk3uu2ee

Urban Traffic Flow Forecast Based on FastGCRNN [article]

Ya Zhang, Mingming Lu, Haifeng Li
2020 arXiv   pre-print
Traffic forecasting is an important prerequisite for the application of intelligent transportation systems in urban traffic networks. The existing works adopted RNN and CNN/GCN, among which GCRN is the state of art work, to characterize the temporal and spatial correlation of traffic flows. However, it is hard to apply GCRN to the large scale road networks due to high computational complexity. To address this problem, we propose to abstract the road network into a geometric graph and build a
more » ... t Graph Convolution Recurrent Neural Network (FastGCRNN) to model the spatial-temporal dependencies of traffic flow. Specifically, We use FastGCN unit to efficiently capture the topological relationship between the roads and the surrounding roads in the graph with reducing the computational complexity through importance sampling, combine GRU unit to capture the temporal dependency of traffic flow, and embed the spatiotemporal features into Seq2Seq based on the Encoder-Decoder framework. Experiments on large-scale traffic data sets illustrate that the proposed method can greatly reduce computational complexity and memory consumption while maintaining relatively high accuracy.
arXiv:2009.08087v1 fatcat:yl34z2zqujdj3fvnk5ulntre4y

Overcoming Catastrophic Forgetting by Soft Parameter Pruning [article]

Jian Peng, Jiang Hao, Zhuo Li, Enqiang Guo, Xiaohong Wan, Deng Min, Qing Zhu, Haifeng Li
2018 arXiv   pre-print
Catastrophic forgetting is a challenge issue in continual learning when a deep neural network forgets the knowledge acquired from the former task after learning on subsequent tasks. However, existing methods try to find the joint distribution of parameters shared with all tasks. This idea can be questionable because this joint distribution may not present when the number of tasks increase. On the other hand, It also leads to "long-term" memory issue when the network capacity is limited since
more » ... ing tasks will "eat" the network capacity. In this paper, we proposed a Soft Parameters Pruning (SPP) strategy to reach the trade-off between short-term and long-term profit of a learning model by freeing those parameters less contributing to remember former task domain knowledge to learn future tasks, and preserving memories about previous tasks via those parameters effectively encoding knowledge about tasks at the same time. The SPP also measures the importance of parameters by information entropy in a label free manner. The experiments on several tasks shows SPP model achieved the best performance compared with others state-of-the-art methods. Experiment results also indicate that our method is less sensitive to hyper-parameter and better generalization. Our research suggests that a softer strategy, i.e. approximate optimize or sub-optimal solution, will benefit alleviating the dilemma of memory. The source codes are available at https://github.com/lehaifeng/Learning_by_memory.
arXiv:1812.01640v1 fatcat:sdi2skmqczhidd56uomwfzbfxy

Heterogeneous Graph Matching Networks [article]

Shen Wang, Zhengzhang Chen, Xiao Yu, Ding Li, Jingchao Ni, Lu-An Tang, Jiaping Gui, Zhichun Li, Haifeng Chen, Philip S. Yu
2019 arXiv   pre-print
Information systems have widely been the target of malware attacks. Traditional signature-based malicious program detection algorithms can only detect known malware and are prone to evasion techniques such as binary obfuscation, while behavior-based approaches highly rely on the malware training samples and incur prohibitively high training cost. To address the limitations of existing techniques, we propose MatchGNet, a heterogeneous Graph Matching Network model to learn the graph
more » ... and similarity metric simultaneously based on the invariant graph modeling of the program's execution behaviors. We conduct a systematic evaluation of our model and show that it is accurate in detecting malicious program behavior and can help detect malware attacks with less false positives. MatchGNet outperforms the state-of-the-art algorithms in malware detection by generating 50% less false positives while keeping zero false negatives.
arXiv:1910.08074v1 fatcat:hgqew4qbwzfvppiy53vsgeynyu

Understanding the Importance of Single Directions via Representative Substitution [article]

Li Chen, Hailun Ding, Qi Li, Zhuo Li, Jian Peng, Haifeng Li
2018 arXiv   pre-print
Understanding the internal representations of deep neural networks (DNNs) is crucal to explain their behavior. The interpretation of individual units, which are neurons in MLPs or convolution kernels in convolutional networks, has been paid much attention given their fundamental role. However, recent research (Morcos et al. 2018) presented a counterintuitive phenomenon, which suggests that an individual unit with high class selectivity, called interpretable units, has poor contributions to
more » ... alization of DNNs. In this work, we provide a new perspective to understand this counterintuitive phenomenon, which makes sense when we introduce Representative Substitution (RS). Instead of individually selective units with classes, the RS refers to the independence of a unit's representations in the same layer without any annotation. Our experiments demonstrate that interpretable units have high RS which are not critical to network's generalization. The RS provides new insights into the interpretation of DNNs and suggests that we need to focus on the independence and relationship of the representations.
arXiv:1811.11053v2 fatcat:gxjn53kymrg75j4zglljhtghja

Scattering-Induced Disk Polarization By Millimeter-Sized Grains [article]

Haifeng Yang, Zhi-Yun Li
2019 arXiv   pre-print
The line of sight lies in the xOz plane and has an inclination angle i with the z axis.  ...  We can see that it lies in the region colored in red with polarization reversal (see above).  ... 
arXiv:1909.08192v1 fatcat:u624j7slq5gjnaba2rf5af4f6e

Scale-Adaptive Adversarial Patch Attack for Remote Sensing Image Aircraft Detection

Mingming Lu, Qi Li, Li Chen, Haifeng Li
2021 Remote Sensing  
With the adversarial attack of convolutional neural networks (CNNs), we are able to generate adversarial patches to make an aircraft undetectable by object detectors instead of covering the aircraft with large camouflage nets. However, aircraft in remote sensing images (RSIs) have the problem of large variations in scale, which can easily cause size mismatches between an adversarial patch and an aircraft. A small adversarial patch has no attack effect on large aircraft, and a large adversarial
more » ... atch will completely cover small aircraft so that it is impossible to judge whether the adversarial patch has an attack effect. Therefore, we propose the adversarial attack method Patch-Noobj for the problem of large-scale variation in aircraft in RSIs. Patch-Noobj adaptively scales the width and height of the adversarial patch according to the size of the attacked aircraft and generates a universal adversarial patch that can attack aircraft of different sizes. In the experiment, we use the YOLOv3 detector to verify the effectiveness of Patch-Noobj on multiple datasets. The experimental results demonstrate that our universal adversarial patches are well adapted to aircraft of different sizes on multiple datasets and effectively reduce the Average Precision (AP) of the YOLOv3 detector on the DOTA, NWPU VHR-10, and RSOD datasets by 48.2%, 23.9%, and 20.2%, respectively. Moreover, the universal adversarial patch generated on one dataset is also effective in attacking aircraft on the remaining two datasets, while the adversarial patch generated on YOLOv3 is also effective in attacking YOLOv5 and Faster R-CNN, which demonstrates the attack transferability of the adversarial patch.
doi:10.3390/rs13204078 fatcat:ehh3ykajijbj7osqcyemm5w5ye

Psittacosis [chapter]

Haifeng Mi, Hongjun Li, Jianan Yu
2015 Radiology of Infectious Diseases: Volume 2  
Psittacosis, also known as ornithosis, is an acute infectious disease caused by Chlamydia psittaci (Cps) and commonly prevails in poultry and other species of bird. Humans infected by Chlamydia psittaci may suffer from unapparent subclinical infection, with symptoms ranging from mild fl ulike illness to severe SARS. As a typical animal-based infectious disease, psittacosis rarely has pulmonary signs but a long illness course, despite its clinical manifestation characterized by severe pulmonary
more » ... esions. Repeated onsets of psittacosis may lead to chronic diseases. Etiology Initially isolated from parrots, Chlamydia psittaci (Cps) is the pathogen of psittacosis. With a diameter of 150-200 nm, the elementary body is ring-shaped and characterized by a narrow protoplasmic margin around the nucleoplasm, a nonglycogen inclusion body and iodine staining negative. Cps develops well in several cell culture systems, among which HeLa cells, Vero cells, and L cells as well as McCoy cells are commonly used. The Cps can also develop in the yolk sac of the chicken embryo. The number of susceptible animals is relatively large, and the laboratory rats are usually used in the animal inoculation. As Cps and Chlamydia trachomatis share the same antigen, both of them cannot be distinguished by the complement fi xation test (CFT). With a weak resistance to the surroundings, Cps can be easily killed by the general chemical disinfectants. It can be inactivated in 48 h at 37 °C, in 10 min at 60 °C, in 24 h with 0.1 % formaldehyde or 0.5 % phenol, and in 30 min with diethyl ether or with ultraviolet radiation. It is resistant to low temperature and can remain infectious for several years if it is kept at -70 °C. Birds which are infected by psittacosis or serve as the pathogen carriers are considered as the source of infection. Currently, more than 140 types of birds are known to contract or carry the pathogen which is mostly found in secretions and feathers. Infections in birds are unapparent and the signs are characteristic. Although most of the infected birds show no or mild symptoms, the pathogens can be excreted for several months. A patient can also become a minor source of infection if he/she excretes pathogens in sputum. Route of Transmission Psittacosis can be transmitted via the respiratory tract. Besides being directly transmitted to humans via droplet, the bacteria can be indirectly transmitted by inhaling an aerosol of infected birds' feces via the respiratory tract. However, according to the reports, few patients experience the onsets without the contact history of birds. Psittacosis is rarely transmitted via direct person-to-person contact. Susceptible Population Populations are generally susceptible and the occurrence has no signifi cant gender difference. It is an epidemic disease all year round. The infection rate is closely related to the frequency of bird contact: parrot and poultry raisers easily contract the disease. Although certain immunity can be acquired after the infection is cured, it is not strong enough to prevent the repeated onsets and the following infection.
doi:10.1007/978-94-017-9876-1_20 fatcat:6iiwotkhbnbgxef2tq6bqkymbm

A Restricted Version of Wythoff's Game

Wen An Liu, Haifeng Li, Bei Li
2011 Electronic Journal of Combinatorics  
E. Duchêne and S. Gravier present the following open problem: In Wythoff's game, each player can either remove at most $R$ tokens from a single heap (i.e. there is an upper bound $R$ on the number of removing tokens), or remove the same number of tokens from both heaps but there is no upper bound on the number of removing tokens. This open problem is investigated and all its P-positions are given.
doi:10.37236/694 fatcat:rtzazey3gfc7rmqo2c2g3acifq
« Previous Showing results 1 — 15 out of 4,129 results