Multimodal Data Fusion And Attack Detection In Recommender Systems,
2020
University of South Florida
Multimodal Data Fusion And Attack Detection In Recommender Systems, Mehmet Aktukmak
USF Tampa Graduate Theses and Dissertations
The commercial platforms that use recommender systems can collect relevant information to produce useful recommendations to the platform users. However, these sources usually contain missing values, imbalanced and heterogeneous data, and noisy observations. Such characteristics render the process of exploiting the information nontrivial, as one should carefully address them during the data fusion process. In addition to the degenerative characteristics, some entries can be fake, i.e., they can be the outcomes of malicious intents to manipulate the system. These entries should be eliminated before incorporation to any recommendation task. Detecting such malicious attacks quickly and accurately and then mitigating them …
Global Context Aware Convolutions For 3d Point Cloud Understanding,
2020
Singapore Management University
Global Context Aware Convolutions For 3d Point Cloud Understanding, Zhiyuan Zhang, Binh-Son Hua, Wei Chen, Yibin Tian, Sai-Kit Yeung
Research Collection School Of Computing and Information Systems
Recent advances in deep learning for 3D point clouds have shown great promises in scene understanding tasks thanks to the introduction of convolution operators to consume 3D point clouds directly in a neural network. Point cloud data, however, could have arbitrary rotations, especially those acquired from 3D scanning. Recent works show that it is possible to design point cloud convolutions with rotation invariance property, but such methods generally do not perform as well as translation-invariant only convolution. We found that a key reason is that compared to point coordinates, rotation-invariant features consumed by point cloud convolution are not as distinctive. …
Uniconv: A Unified Conversational Neural Architecture For Multi-Domain Task-Oriented Dialogues,
2020
Singapore Management University
Uniconv: A Unified Conversational Neural Architecture For Multi-Domain Task-Oriented Dialogues, Hung Le, Doyen Sahoo, Chenghao Liu, Nancy F. Chen, Steven C. H. Hoi
Research Collection School Of Computing and Information Systems
Building an end-to-end conversational agent for multi-domain task-oriented dialogues has been an open challenge for two main reasons. First, tracking dialogue states of multiple domains is non-trivial as the dialogue agent must obtain complete states from all relevant domains, some of which might have shared slots among domains as well as unique slots specifically for one domain only. Second, the dialogue agent must also process various types of information across domains, including dialogue context, dialogue states, and database, to generate natural responses to users. Unlike the existing approaches that are often designed to train each module separately, we propose “UniConv" …
Law Library Blog (November 2020): Legal Beagle's Blog Archive,
2020
Roger Williams University
Law Library Blog (November 2020): Legal Beagle's Blog Archive, Roger Williams University School Of Law
Law Library Newsletters/Blog
No abstract provided.
Extending The Functional Subnetwork Approach To A Generalized Linear Integrate-And-Fire Neuron Model,
2020
West Virginia University
Extending The Functional Subnetwork Approach To A Generalized Linear Integrate-And-Fire Neuron Model, Nicholas Szczecinski, Roger Quinn, Alexander J. Hunt
Mechanical and Materials Engineering Faculty Publications and Presentations
Engineering neural networks to perform specific tasks often represents a monumental challenge in determining network architecture and parameter values. In this work, we extend our previously-developed method for tuning networks of non-spiking neurons, the “Functional subnetwork approach” (FSA), to the tuning of networks composed of spiking neurons. This extension enables the direct assembly and tuning of networks of spiking neurons and synapses based on the network’s intended function, without the use of global optimization ormachine learning. To extend the FSA, we show that the dynamics of a generalized linear integrate and fire (GLIF) neuronmodel have fundamental similarities to those of …
Machine Learning Integrated Design For Additive Manufacturing,
2020
Singapore Management University
Machine Learning Integrated Design For Additive Manufacturing, Jingchao Jiang, Yi Xiong, Zhiyuan Zhang, David W. Rosen
Research Collection School Of Computing and Information Systems
For improving manufacturing efficiency and minimizing costs, design for additive manufacturing (AM) has been accordingly proposed. The existing design for AM methods are mainly surrogate model based. Due to the increasingly available data nowadays, machine learning (ML) has been applied to medical diagnosis, image processing, prediction, classification, learning association, etc. A variety of studies have also been carried out to use machine learning for optimizing the process parameters of AM with corresponding objectives. In this paper, a ML integrated design for AM framework is proposed, which takes advantage of ML that can learn the complex relationships between the design and …
Bist: Bi-Directional Spatio-Temporal Reasoning For Video-Grounded Dialogues,
2020
Singapore Management University
Bist: Bi-Directional Spatio-Temporal Reasoning For Video-Grounded Dialogues, Hung Le, Doyen Sahoo, Nancy F. Chen, Steven C. H. Hoi
Research Collection School Of Computing and Information Systems
Video-grounded dialogues are very challenging due to (i) the complexity of videos which contain both spatial and temporal variations, and (ii) the complexity of user utterances which query different segments and/or different objects in videos over multiple dialogue turns. However, existing approaches to video-grounded dialogues often focus on superficial temporal-level visual cues, but neglect more fine-grained spatial signals from videos. To address this drawback, we propose Bi-directional Spatio-Temporal Learning (BiST), a vision-language neural framework for high-resolution queries in videos based on textual cues. Specifically, our approach not only exploits both spatial and temporal-level information, but also learns dynamic information diffusion …
Communicating Uncertain Information From Deep Learning Models In Human Machine Teams,
2020
Missouri University of Science and Technology
Communicating Uncertain Information From Deep Learning Models In Human Machine Teams, Harishankar V. Subramanian, Casey I. Canfield, Daniel Burton Shank, Luke Andrews, Cihan H. Dagli
Engineering Management and Systems Engineering Faculty Research & Creative Works
The role of human-machine teams in society is increasing, as big data and computing power explode. One popular approach to AI is deep learning, which is useful for classification, feature identification, and predictive modeling. However, deep learning models often suffer from inadequate transparency and poor explainability. One aspect of human systems integration is the design of interfaces that support human decision-making. AI models have multiple types of uncertainty embedded, which may be difficult for users to understand. Humans that use these tools need to understand how much they should trust the AI. This study evaluates one simple approach for communicating …
Finite-Time State Estimation For An Inverted Pendulum Under Input-Multiplicative Uncertainty,
2020
Embry-Riddle Aeronautical University
Finite-Time State Estimation For An Inverted Pendulum Under Input-Multiplicative Uncertainty, Sergey V. Drakunov, William Mackunis, Anu Kossery Jayaprakash, Krishna Bhavithavya Kidambi, Mahmut Reyhanoglu
Publications
A sliding mode observer is presented, which is rigorously proven to achieve finite-time state estimation of a dual-parallel underactuated (i.e., single-input multi-output) cart inverted pendulum system in the presence of parametric uncertainty. A salient feature of the proposed sliding mode observer design is that a rigorous analysis is provided, which proves finite-time estimation of the complete system state in the presence of input-multiplicative parametric uncertainty. The performance of the proposed observer design is demonstrated through numerical case studies using both sliding mode control (SMC)- and linear quadratic regulator (LQR)-based closed-loop control systems. The main contribution presented here is the rigorous …
Espade: An Efficient And Semantically Secure Shortest Path Discovery For Outsourced Location-Based Services,
2020
Montclair State University
Espade: An Efficient And Semantically Secure Shortest Path Discovery For Outsourced Location-Based Services, Bharath K. Samanthula, Divyadharshini Karthikeyan, Boxiang Dong, K. Anitha Kumari
Department of Computer Science Faculty Scholarship and Creative Works
With the rapid growth of smart devices and technological advancements in tracking geospatial data, the demand for Location-Based Services (LBS) is facing a constant rise in several domains, including military, healthcare and transportation. It is a natural step to migrate LBS to a cloud environment to achieve on-demand scalability and increased resiliency. Nonetheless, outsourcing sensitive location data to a third-party cloud provider raises a host of privacy concerns as the data owners have reduced visibility and control over the outsourced data. In this paper, we consider outsourced LBS where users want to retrieve map directions without disclosing their location information. …
Research On Optimal Configuration Design Method Of Stewart Platform,
2020
Shanghai Electro-Mechanical Engineering Institute, Shanghai 201109, China;
Research On Optimal Configuration Design Method Of Stewart Platform, Xuwei Fan, Lili Yang, Cheng Yu, Xiaoning Zhou, Yexin Zhang
Journal of System Simulation
Abstract: Based on the structural singularity and configuration singularity of parallel mechanism, a safety mechanism design scheme of the Stewart platform is proposed to improve the workspace efficiency and engineering practicability. Taking the Stewart platform without any particularity as the research object, and considering the singular constraints of the structure, a dexterity index is proposed to achieve the optimization of the structural parameters. Analyzing and constructing the kinematics model of the Stewart platform, analyzing the singularities of the configuration bifurcations of 16 typical extreme poses, a secure workspace verification algorithm is proposed to make the whole workspace free of singularity. …
Research On Adaptive Routing Algorithm For Wireless Weak-Connection Network,
2020
1. Northwest Institutes of Advanced Technology, Xi'an Technological University, Xi'an 710021, China; ;2. College of Electronic Information Engineering, Xi'an Technological University, Xi'an 710021, China; ;
Research On Adaptive Routing Algorithm For Wireless Weak-Connection Network, Hua Xiang, Hongjuan Yao, Wang Hai, Wang Zhao, Jietao Zhang, Lili Shu
Journal of System Simulation
Abstract: The wireless weak-connected network has the characteristics of long delay, high dynamic topology, and unstable links. With the lack of continuity from the source end to the destination end of network connection, in order to solve the problem of communication difficulty, the intelligence and adaptability of Physarum polycephalum are introduced, and the adaptive wireless weak-connected network routing algorithm is proposed. A wireless weak-connected network model is build and the mathematical relationships of link capacity is deduced. The next-hop selection strategy and optimal routing strategy is designed to achieve the best-effort delivery of data in wireless weak-connected network environmrnt. Simulation …
Visual Feedback Fuzzy Control For A Robot Manipulator Based On Svr Learning,
2020
School of Mechatronics and Automation, Shanghai University, Shanghai 200444, China;
Visual Feedback Fuzzy Control For A Robot Manipulator Based On Svr Learning, Xianxia Zhang, Jinqiang Zhang, Zhiyuan Li, Shiwei Ma, Banghua Yang
Journal of System Simulation
Abstract: A fuzzy controller based on SVR learning is proposed for uncalibrated robot visual servoing. In this paper, a fuzzy controller is used to directly construct the nonlinear mapping between image features and robot joint motion. The fuzzy basis function of the fuzzy controller is taken as the kernel function of an SVR and the equivalent relationship between the SVR and the fuzzy controller is established. The learned support vector from the SVR is used as the rule of the fuzzy controller. Since all rules are learned from the data, there is no need to manually design the rules. …
Gru-Based Car-Following Behavior Simulation Model,
2020
1. School of Computer Science, Xi'an University of Technology, Xi'an 710048, China; ;
Gru-Based Car-Following Behavior Simulation Model, Fei Rong, Liu Fang, Xie Guo, Hei Xinhong, Shasha Li, Hu Bo
Journal of System Simulation
Abstract: The accuracy of acceleration prediction can be effectively improved by the driver's memory in car-following behavior. A new car-following model based on the General Motors (GM) and the gate control unit (GRU) is proposed. The car-following data between small vehicles with similar driving behavior are obtained by data preprocessing. The established model is calibrated by the car-following data, and the optimal parameters and structure of the model are determined. According to car-following characteristics, the effectiveness of model is verified by simulation. It is confirmed that the model has high robustness and improved simulation accuracy comparing with the traditional models.
Estimation Of Space-Time Of Urban Building Population Based On Mobile Phone Big Data,
2020
Beijing Information Science Technology University Beijing Key Laboratory of High Dynamic Navigation Technology, Beijing 100101, China;
Estimation Of Space-Time Of Urban Building Population Based On Mobile Phone Big Data, Hu Yang, Xiaoyong Zhang, Xiao Di
Journal of System Simulation
Abstract: With the acceleration of the urbanization process, mastering the population distribution on a fine scale is of great significance for urban disaster assessment, emergency response management and public resource allocation. The rapid development of the Internet and the popularity of smartphones have prompted mobile phones to become the sensors of human activity. A method of urban building population estimation based on mobile phone big data is proposed. The method analyzes the crowd activity law of buildings with different functions based on mobile phone positioning big data in typical areas, calculates the population capacity of different functional buildings, and …
Simulation Research On Optimization Of Agv Charging Strategy For Automated Terminal,
2020
1. Dalian Neusoft University of Information, Dalian 116023, China; ;
Simulation Research On Optimization Of Agv Charging Strategy For Automated Terminal, Danlan Xie, Guo Di, Ji Yuan, Zhipeng Gao
Journal of System Simulation
Abstract: As the automated container terminal is the development trend of terminal, AGV (automated guided vehicle) becomes the most widely used horizontal transportation tool and it is important to make its reasonable charging strategy. Aiming at the shortcomings of the current AGV, a charging strategy of offline charging being primary and online charging being auxiliary is proposed. In order to solve the problem of location selection of online charging station, a quick method of selecting effective stations by using heat zone map is proposed. Through a large number of simulation experiments using this strategy, the fact that the number …
Global Optimization Method Based On Consensus Particle Swarm Optimization,
2020
1. School of Electrical Engineering, University of Ji'nan, Ji'nan 250022, China; ;
Global Optimization Method Based On Consensus Particle Swarm Optimization, Zhanwen Lu, Xingong Cheng, Yongfeng Zhang
Journal of System Simulation
Abstract: According to the characteristics of particle swarm optimization (PSO) and efficient global optimization algorithm (EGO), a global black box optimization algorithm based on consensus particle swarm optimization and local surrogate model (CPSO-LSM) is proposed. The algorithm fixes the period of the PSO algorithm to group the particles and stops after the particles reach a consensus. The high-quality sub-regions around each group of particles are used as the modeling area of the surrogate model, and the high-quality optimal solution or global optimal solution is obtained by comparing the optimal values of each region. It can not only avoid the complex …
Natural Computing Method Based On Lle Dimension Reduction,
2020
College of Computer Science and Information Engineering, Harbin Normal University, Harbin 150025, China;
Natural Computing Method Based On Lle Dimension Reduction, Luyao Zhang, Weidong Ji, Cheng Hao
Journal of System Simulation
Abstract: In the natural computing method, the appearance of high-dimensional problem can make some existing optimization algorithms avoid falling into local optimum, but it makes the performance of the algorithm worse and the running time longer. On the basis of traditional natural calculation method, a natural calculation method based on LLE(Local Linear Embedding) algorithm is proposed, which analyzes the value of neighbor particle k and dimension d, and makes the algorithm get better optimization effect after dimension reduction. In the process, a small bias s is added to the data after dimension reduction to increase the diversity of …
Research On Fuzzy Flexible Job Shop Scheduling Problem Based On Hybrid Qpso,
2020
Engineering Research Center of Internet of Things Technology Applications Ministry of Education, Wuxi 214122, China;
Research On Fuzzy Flexible Job Shop Scheduling Problem Based On Hybrid Qpso, Junxuan Li, Wang Yan, Zhicheng Ji
Journal of System Simulation
Abstract: To solve the flexible job shop scheduling problem of uncertain processing time, triangular fuzzy numbers are used to characterize the relevant time parameters and a Hybrid Quantum Particle Swarm Optimization (HQPSO) is proposed. On the basis of making full use of the global search capability of Quantum Particle Swarm Optimization, the search efficiency is increased by designing a boundary repair strategy and a cooperative update strategy. Meanwhile, the cross-operator and path relinking technique are used to in the operation sequence mapped by the excellent particles, which makes up the disadvantages of the insufficient ability of deeply exploration of …
Adaptive Control Method Of High Proportion Distributed Generation Connected To Distribution Network,
2020
1. Beijing Information Science and Technology University Beijing 100192, China; ;
Adaptive Control Method Of High Proportion Distributed Generation Connected To Distribution Network, Weijie Dong, Keyan Liu, Yilong Wang, Xiaozhong Li, Yang Hao
Journal of System Simulation
Abstract: Focus on the instability of the reactive power output precision caused by the voltage drop and loss of DG lines being switched to grid and the low output frequency and voltage caused by power deficiency of DG islanding, an improved droop control strategy is proposed to ensure the running stability of Active Distribution Network during DG connecting to network. The DG model applies the feedback regulation of f-U droop output power and voltage drop compensator. The control system is applied to the quasi-synchronization grid-connected model, island mode and the modes switching process of an active distribution network's domain transient …
