Open Access. Powered by Scholars. Published by Universities.®
Numerical Analysis and Scientific Computing Commons™
Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Engineering (311)
- Artificial Intelligence and Robotics (302)
- Operations Research, Systems Engineering and Industrial Engineering (289)
- Computer Engineering (287)
- Systems Science (281)
-
- Databases and Information Systems (37)
- Social and Behavioral Sciences (19)
- Applied Mathematics (14)
- Data Science (14)
- Theory and Algorithms (14)
- Physics (12)
- Business (11)
- Software Engineering (11)
- Numerical Analysis and Computation (9)
- Statistics and Probability (9)
- Chemistry (7)
- Electrical and Computer Engineering (7)
- Graphics and Human Computer Interfaces (7)
- Probability (7)
- Statistical Models (7)
- Applied Statistics (6)
- Medicine and Health Sciences (6)
- Programming Languages and Compilers (6)
- Public Affairs, Public Policy and Public Administration (6)
- Biomedical Engineering and Bioengineering (5)
- Categorical Data Analysis (5)
- Econometrics (5)
- Institution
-
- China Simulation Federation (281)
- Singapore Management University (57)
- University of Arkansas, Fayetteville (7)
- Central Bank of Nigeria (4)
- Embry-Riddle Aeronautical University (4)
-
- City University of New York (CUNY) (3)
- Claremont Colleges (2)
- Kennesaw State University (2)
- LSU New Orleans (2)
- Louisiana State University (2)
- Mississippi State University (2)
- University at Albany, State University of New York (2)
- University of Kentucky (2)
- University of Montana (2)
- University of South Alabama (2)
- American University in Cairo (1)
- Bowling Green State University (1)
- Chapman University (1)
- Clemson University (1)
- Colby College (1)
- Dartmouth College (1)
- Duquesne University (1)
- Illinois State University (1)
- James Madison University (1)
- Loyola University Chicago (1)
- Murray State University (1)
- Nova Southeastern University (1)
- Old Dominion University (1)
- Purdue University (1)
- Rose-Hulman Institute of Technology (1)
- Keyword
-
- Deep learning (14)
- Machine Learning (7)
- Deep reinforcement learning (6)
- Energy consumption (5)
- Simulation (5)
-
- Digital twin (4)
- Machine learning (4)
- Path planning (4)
- Resource allocation (4)
- Trajectory planning (4)
- Virtual reality (4)
- Attention mechanism (3)
- COVID-19 (3)
- China (3)
- Cloud platform (3)
- Convolutional neural network (3)
- Deep Learning (3)
- Edge computing (3)
- GPU (3)
- Machine vision (3)
- Mixed traffic flow (3)
- Modeling (3)
- Multi-objective optimization (3)
- Reinforcement learning (3)
- Sensitivity analysis (3)
- Sentiment analysis (3)
- Simulation model (3)
- Social force model (3)
- System dynamics (3)
- Time series (3)
- Publication
-
- Journal of System Simulation (281)
- Research Collection School Of Computing and Information Systems (54)
- Computer Science and Computer Engineering Undergraduate Honors Theses (5)
- CBN Journal of Applied Statistics (JAS) (4)
- Theses and Dissertations (3)
-
- Dissertations, Theses, and Capstone Projects (2)
- Doctoral Dissertations and Master's Theses (2)
- Graduate Student Theses, Dissertations, & Professional Papers (2)
- Graduate Theses and Dissertations (2)
- Graduate Theses and Dissertations (2019 - present) (2)
- Honors Theses (2)
- LSU New Orleans Theses and Dissertations (2)
- Legacy Theses & Dissertations (2009 - 2024) (2)
- All Dissertations (1)
- Annual Symposium on Biomathematics and Ecology Education and Research (1)
- Beyond: Undergraduate Research Journal (1)
- CMC Senior Theses (1)
- Computer Science: Faculty Publications and Other Works (1)
- Conference Papers (1)
- Dartmouth College Master’s Theses (1)
- Dissertations and Theses Collection (Open Access) (1)
- Electrical Engineering Theses (1)
- Electrical and Computer Engineering Publications (1)
- Electronic Theses and Dissertations (1)
- Engineering Faculty Articles and Research (1)
- Holland Computing Center: Faculty Publications (1)
- Honors College Theses (1)
- Honors Projects (1)
- Honors Theses and Capstones (1)
- International Journal for Research in Education (1)
- Publication Type
- File Type
Articles 31 - 60 of 405
Full-Text Articles in Numerical Analysis and Scientific Computing
A Recommendation On How To Teach K-Means In Introductory Analytics Courses, Manoj Thulasidas
A Recommendation On How To Teach K-Means In Introductory Analytics Courses, Manoj Thulasidas
Research Collection School Of Computing and Information Systems
We teach K-Means clustering in introductory data analytics courses because it is one of the simplest and most widely used unsupervised machine learning algorithms. However, one drawback of this algorithm is that it does not offer a clear method to determine the appropriate number of clusters; it does not have a built-in mechanism for K selection. What is usually taught as the solution for the K Selection problem is the so-called elbow method, where we look at the incremental changes in some quality metric (usually, the sum of squared errors, SSE), trying to find a sudden change. In addition to …
An Efficient Annealing-Assisted Differential Evolution For Multi-Parameter Adaptive Latent Factor Analysis, Qing Li, Guansong Pang, Mingsheng Shang
An Efficient Annealing-Assisted Differential Evolution For Multi-Parameter Adaptive Latent Factor Analysis, Qing Li, Guansong Pang, Mingsheng Shang
Research Collection School Of Computing and Information Systems
A high-dimensional and incomplete (HDI) matrix is a typical representation of big data. However, advanced HDI data analysis models tend to have many extra parameters. Manual tuning of these parameters, generally adopting the empirical knowledge, unavoidably leads to additional overhead. Although variable adaptive mechanisms have been proposed, they cannot balance the exploration and exploitation with early convergence. Moreover, learning such multi-parameters brings high computational time, thereby suffering gross accuracy especially when solving a bilinear problem like conducting the commonly used latent factor analysis (LFA) on an HDI matrix. Herein, an efficient annealing-assisted differential evolution for multi-parameter adaptive latent factor analysis …
Mining Competitively-Priced Bundle Configurations, Ezekiel Ong Young, Hady W. Lauw
Mining Competitively-Priced Bundle Configurations, Ezekiel Ong Young, Hady W. Lauw
Research Collection School Of Computing and Information Systems
We examine the bundle configuration problem in the presence of competition. Given a competitor's bundle configuration and pricing, we determine what to bundle together, and at what prices, to maximize the target firm's revenue. We highlight the difficulty in pricing bundles and propose a scalable alternative and an efficient search heuristic to refine the approximate prices. Furthermore, we extend the heuristics proposed by previous work to accommodate the presence of a competitor. We analyze the effectiveness of our proposed models through experimentation on real-life ratings-based preference data.
Pacific: Towards Proactive Conversational Question Answering Over Tabular And Textual Data In Finance, Yang Deng, Wenqiang Lei, Wenxuan Zhang, Wai Lam, Tat-Seng Chua
Pacific: Towards Proactive Conversational Question Answering Over Tabular And Textual Data In Finance, Yang Deng, Wenqiang Lei, Wenxuan Zhang, Wai Lam, Tat-Seng Chua
Research Collection School Of Computing and Information Systems
To facilitate conversational question answering (CQA) over hybrid contexts in finance, we present a new dataset, named PACIFIC. Compared with existing CQA datasets, PACIFIC exhibits three key features: (i) proactivity, (ii) numerical reasoning, and (iii) hybrid context of tables and text. A new task is defined accordingly to study Proactive Conversational Question Answering (PCQA), which combines clarification question generation and CQA. In addition, we propose a novel method, namely UniPCQA, to adapt a hybrid format of input and output content in PCQA into the Seq2Seq problem, including the reformulation of the numerical reasoning process as code generation. UniPCQA performs multi-task …
A Wind Turbine Fault Diagnosis Method Based On Siamese Deep Neural Network, Jiarui Liu, Guotian Yang, Xiaowei Wang
A Wind Turbine Fault Diagnosis Method Based On Siamese Deep Neural Network, Jiarui Liu, Guotian Yang, Xiaowei Wang
Journal of System Simulation
Abstract: In order to effectively extract the fault features of time series data in supervisory control and data acquisition (SCADA), considering the advantages of one-dimensional convolutional neural network (1-D CNN) for extracting local time series features and the advantages of long-term memory (LSTM) which can extract long-term dependent features, a method for fault diagnosis of wind turbines based on 1-D CNN-LSTM is proposed. To solve the problem of the scarcity of fault samples of wind turbines based on the siamese network architecture, a wind fault diagnosis method based on siamese 1-D CNN-LSTM is proposed. The proposed siamese 1-D CNN-LSTM …
Transmission Line Insulator Recognition Based On Artificial Images Data Expansion, Yaru Wang, Kai Yang, Yongjie Zhai, Congbin Guo, Wenqing Zhao, Jie Su
Transmission Line Insulator Recognition Based On Artificial Images Data Expansion, Yaru Wang, Kai Yang, Yongjie Zhai, Congbin Guo, Wenqing Zhao, Jie Su
Journal of System Simulation
Abstract: Deep learning method has developed rapidly in the field of computer vision, but relies on a large quantities of training data. In the task of transmission line insulator automatic detection, problems such as insufficient number of aerial insulator images and poor diversity affect the accuracy of insulator recognition. An artificial insulator images data expansion method is proposed. Artificial insulator images are created by modeling software, and a compensation network is constructed. The artificial images are compensated and optimized by compensation network, and the aerial insulator image data set is expanded by the compensated artificial insulator images. The insulator recognition …
Contribution Rate Calculation Method To System-Of-Systems Based On Interval-Valued Intuitionistic Fuzzy Number Anp, Zejian Ding, Songtao Sun, Zhiwen He, Fei Liu
Contribution Rate Calculation Method To System-Of-Systems Based On Interval-Valued Intuitionistic Fuzzy Number Anp, Zejian Ding, Songtao Sun, Zhiwen He, Fei Liu
Journal of System Simulation
Abstract: Contribution rate to system-of-systems (CRSoS) is mainly used to measure the contribution of an equipment to system of systems (SoS) in system construction. In order to solve some problems in the calculation of CRSoS, a multi-level equipment indicator architecture of "task-ability-indicator- equipment" is proposed. At the same time, considering the characteristics of the equipment indicator architecture, ANP (analytic network process) and IVIFN (interval-valued intuitionistic fuzzy number), a IVIF-ANP calculation method is proposed to obtain more accurate CRSoS. Experiments show that this method can not only solve the problem of the calculation formula of CRSoS, but also obtain more …
Simulation And Effectiveness Evaluation System For Joint Delivery Mission Planning Of Airlift Fleets, Guochen Wang
Simulation And Effectiveness Evaluation System For Joint Delivery Mission Planning Of Airlift Fleets, Guochen Wang
Journal of System Simulation
Abstract: Airlift fleet plays an important role in modern war. Compared with other countries such as the USA and Russia, China's airlift fleet still has obvious shortcomings and deficiencies. To analysis and optimize the future fleet alternatives, a software tool is established with the modules of model construction and management, scenarios editing, mission planning, simulation deduction, effectiveness analysis. This tool mainly focuses on the interactive relationship between the transport aircraft and cargo, airport and so on, as well as the cooperative relationship of different types of aircraft, which can realize the functions of automatic generation of loading schemes, automatic planning …
Bilevel Distributed Optimal Dispatch Of Active Distribution Network With Multi-Microgrids, Yongjun Lin, Xin Chen, Kai Yang, Shanshan Zhou, Qingfei Bai
Bilevel Distributed Optimal Dispatch Of Active Distribution Network With Multi-Microgrids, Yongjun Lin, Xin Chen, Kai Yang, Shanshan Zhou, Qingfei Bai
Journal of System Simulation
Abstract: With continuous increase of the penetration proportion of renewable energy in the distribution network, the traditional centralized dispatching is facing problems such as high pressure of power flow calculation and difficulty in recycling renewable energy, which makes it difficult to guarantee the operation quality of the system. A distributed optimal two-layer scheduling method for active distribution networks with multiple micro-grids is proposed. The upper-level aims to minimize the loss of regional distribution network, the second-order conical relaxation method and synchronous ADMM (alternating direction method of multipliers) algorithm are used to solve the scheduling instructions of the micro-grid connection lines. …
Research On Key Technology Of Uavs Autonomous Landing Based On Relative Precise Point Position, Guohua Kang, Teng Zhao, Yao Fu, Weizheng Xu, Jianyu Wei, Yuhuan Qiu, Junfeng Wu
Research On Key Technology Of Uavs Autonomous Landing Based On Relative Precise Point Position, Guohua Kang, Teng Zhao, Yao Fu, Weizheng Xu, Jianyu Wei, Yuhuan Qiu, Junfeng Wu
Journal of System Simulation
Abstract: In complex sea conditions with wind and waves, the relative motion between unmanned aerial vehicles (UAVs) requiring autonomous landing and ships is highly uncertain. In order to improve the accuracy of relative positioning and control during autonomous landing, and to ensure the safety and reliability of autonomous landing, a relative precise point positioning (RPPP) technique based on differential tropospheric error is proposed. The technology only relies on data link and carrier satellite positioning receiver to eliminate the same error of satellite positioning in the same environment and obtain accurate relative positioning. The combination of proportional navigation and linear quadratic …
Robust Optimal Configuration Of Pv-Energy Storage In Industrial Parks Considering The Uncertainty Of Photovoltaics, Guiting Xue, Boya Shan, Ti Wang, Xiao Wang, Wei Xing, Weiqing Sun
Robust Optimal Configuration Of Pv-Energy Storage In Industrial Parks Considering The Uncertainty Of Photovoltaics, Guiting Xue, Boya Shan, Ti Wang, Xiao Wang, Wei Xing, Weiqing Sun
Journal of System Simulation
Abstract: Research on using rooftop resources in industrial parks to develop photovoltaic projects and reasonable configuration of energy storage will help improve the park's energy economy. To obtain the optimal PV-storage configuration scheme, an industrial park with three types of load demand, namely, cold, heat and electricity, is selected, and a robust optimization allocation model of park PV-storage is established with optimal operating profit as the objective function, considering the increased cost of power purchase caused by the PV uncertainty. The model uses the box uncertainty set in robust optimization for PV intensity, and linearizes the model using pairwise …
Evacuation Model Considering The Restricted View Of The Sign, Yinghua Song, Zheqian Zhang, Feizhou Huo, Danhui Fang
Evacuation Model Considering The Restricted View Of The Sign, Yinghua Song, Zheqian Zhang, Feizhou Huo, Danhui Fang
Journal of System Simulation
Abstract: In order to study the influence of restricted vision on the process of pedestrian evacuation, a cellular automata model of pedestrian evacuation under restricted vision is established. In the model, the evacuation space is divided into three different areas according to the field of view radius, pedestrians have different ways of moving in different areas. Different income parameters are defined to calculate the pedestrian movement income matrix and determine the target position of the pedestrian in the next time step. An evacuation scene is established to simulate the initial density of different pedestrians, the change of the field of …
Simulation Of O2o Platform Transaction Considering The Constituted Information, Wen Zheng, Ke Shi, Jingyi Zhu
Simulation Of O2o Platform Transaction Considering The Constituted Information, Wen Zheng, Ke Shi, Jingyi Zhu
Journal of System Simulation
Abstract: In the transactions concluded with the help of APP, the information intervenes into the transaction process in the form of constituted information. The constituted information is classified into three degrees in a two-sided market. By introducing the three degrees of the constituted information: information acceptance (IA), information diversity (ID) and information loss (IL), the Swarm Class Library and the interactive Agents are designed to construct a systematic O2O (online to offline)platform transaction model, which encapsulates the Consumer/Seller/PlatformAgents. The constituted information acts as the systematic conditions of invoking and judging, and the graphical user interface (GUI) outputs the …
Simulation And Experiment Of An Intelligent Control Model For The Cleaning Of A Rice-Wheat Combine Harvester, Qing Jiang, Rujing Wang
Simulation And Experiment Of An Intelligent Control Model For The Cleaning Of A Rice-Wheat Combine Harvester, Qing Jiang, Rujing Wang
Journal of System Simulation
Abstract: Modeling and simulation of cleaning intelligent control according to the changes of cleaning loss rate and trash content rate is the focus of research and hot issues of intelligent control of rice-wheat combine harvester. A method for acquiring knowledge of the experts' experience and knowledge is constructed based on the flow chart of cleaning control, and a cleaning intelligent control knowledge base for the intelligent control of the field operation environment based on the production rule is proposed. Based on the principle of human-simulating intelligent control, a knowledge inference algorithm for adaptive selection of cleaning regulation strategy is …
Research On Mobile Edge Computing Resource Allocation With Energy Harvesting Device, Changyun Li, Jianbo Li, Xi Xu, Tingli Li
Research On Mobile Edge Computing Resource Allocation With Energy Harvesting Device, Changyun Li, Jianbo Li, Xi Xu, Tingli Li
Journal of System Simulation
Abstract: In order to solve the problem of computing resource allocation of mobile edge computing system with energy gathering ability, an algorithm based on Lyapunov greed optimization (LGO) is proposed. This paper presents a dynamic optimization problem to minimize the combined cost of time delay and energy consumption of mobile devices under the gradual convergence of equipment battery power. Using Lyapunov dynamic optimization theory, the optimization problem is decomposed into three sub-problems of optimal local execution, unloading execution and energy harvesting for each time slot, and the optimal solution of the sub-problems is obtained by linear programming. By selecting the …
Multi-Uav Trajectory Planning Based On Adaptive Segmented Potential Field Method, Guangjian Tian, Jiyang Dai, Jin Ying, Ning Wang
Multi-Uav Trajectory Planning Based On Adaptive Segmented Potential Field Method, Guangjian Tian, Jiyang Dai, Jin Ying, Ning Wang
Journal of System Simulation
Abstract: To solve the problems that the traditional artificial potential field method is prone to fall into the local extreme value, target unreachability and excessive curvature of the planned trajectory curvature in the application of UAV trajectory planning, on the basis of the layered potential field method, a method of adding a second local attractive field at the target point and an attractive set composed of the target attractive field is proposed. This method overcomes the defects of unreachable targets and easy falling into local extremes. In addition, a piecewise function is introduced into the original layered potential field method, …
Nonlinear System Identification Based On Combined Signal Sources, Tian Zheng, Feng Li, Naibao He, Ya Gu
Nonlinear System Identification Based On Combined Signal Sources, Tian Zheng, Feng Li, Naibao He, Ya Gu
Journal of System Simulation
Abstract: Aiming at the interference of noise in the nonlinear system, the identification modeling method of the neuro-fuzzy Hammerstein output error nonlinear system is considered. The combined signal sources are used to realize the parameter identification separation of the linear block and the nonlinear block. The correlation analysis method and the recursive least square identification method based on auxiliary model technique are derived to estimate the parameters of dynamic linear block and nonlinear block, which can effectively suppress the interference of system output noise. Compared with least square algorithm, polynomial model and multi-innovation method, the simulation results demonstrate that the …
Research On Network Public Opinion Transmission Mechanism Of Inversion Event Based On Integrating Improved Sir Model, Jianrong Tang, Jiatong Bao
Research On Network Public Opinion Transmission Mechanism Of Inversion Event Based On Integrating Improved Sir Model, Jianrong Tang, Jiatong Bao
Journal of System Simulation
Abstract: In order to identify the spreading rules of rumors in vicious news reversal events and make more targeted guiding decisions, a short-term prediction model is proposed to simulate the spread of virus information. This paper improves the traditional susceptible infected removed (SIR) model and solves the problem that the conversion rate is fixed and single due to the limitation of Markov chain when it is combined with systems dynamics (SD) model. The data is validated with the example of "asthmatic girls" . The results show that the model not only effectively simulates the crisis of public opinion communication in …
A Real-Time Ultrasound Simulation Platform Using Ray Tracing And Its Integration With Virtual Reality, Bo Peng, Qiang Wang, Ruibing Qing, Lixue Yin, Jingfeng Jiang
A Real-Time Ultrasound Simulation Platform Using Ray Tracing And Its Integration With Virtual Reality, Bo Peng, Qiang Wang, Ruibing Qing, Lixue Yin, Jingfeng Jiang
Journal of System Simulation
Abstract: In order to further improve the efficacy of ultrasound training and reduce the cost. An ultrasound training system that is integrated with a VR environment is developed. The main contribution of this study is to incorporate the Ray-tracing based ultrasound image simulation approach into a virtual reality environment, taking advantage of immersive VR experience for medical ultrasound training. The simulated ultrasound images obtained by the proposed method are then compared to images that are simulated using a generative adversarial network (GAN) technique and Field II ultrasound simulator. The data show that the ultrasound simulator can produce high-quality simulated …
A Two-Layer Network Propagation Model Of Awareness Diffusion And Seir Epidemic, Yurong Song, Yulin Bao, Ruqi Li
A Two-Layer Network Propagation Model Of Awareness Diffusion And Seir Epidemic, Yurong Song, Yulin Bao, Ruqi Li
Journal of System Simulation
Abstract: In order to understand the transmission characteristics of epidemics similar to COVID-19 (coronavirus disease 2019) with obvious expose period, a two-layer network transmission model considering time-varying factors is proposed to make corresponding predictions and measures. The UAU (unaware-aware-unaware) information transmission model is used to represent the diffusion process of conscious information about epidemic. In the underlying network, the susceptible-exposed-infected- recovered (SEIR) epidemic-like transmission model with latent state is used to describe the epidemic transmission process affected by conscious information. The MMCA (microscopic Markov chain approach) is used to deduce the transmission threshold of epidemics diseases. By analyzing the key …
Design And Simulation-Based Evaluation Of Taxiway Operation Scheme For Multi-Runway Airport Maneuvering Area, Xinping Zhu, Chuan Xu, Jingjing Qu, Tingwen Su
Design And Simulation-Based Evaluation Of Taxiway Operation Scheme For Multi-Runway Airport Maneuvering Area, Xinping Zhu, Chuan Xu, Jingjing Qu, Tingwen Su
Journal of System Simulation
Abstract: The operational scheme of the taxiway system is very important to promote the efficient utilization of airfield resources in multi-runway airports. The design and simulation evaluation methods of the taxiway operation scheme for multi-runway airports are studied. The design principles of "fixed, unidirectional, compliant and circular" taxiway operation scheme and the design paradigm of the operation scheme are presented, and the concepts of taxiway space occupancy index and potential conflict index are proposed. Using Haikou Meilan International Airport as the application scenario, the corresponding optimization scheme of the taxiway system in the maneuvering area is given based on …
Multi-Market Coupling Trading Simulation Of Electricity Green Certificate And Excess Consumption Under New Renewable Portfolio Standard, Peng Wang, Xiaohua Song, Haowen Yang, Xiaoying Zhai, Jingjing Han, Liwei Ju
Multi-Market Coupling Trading Simulation Of Electricity Green Certificate And Excess Consumption Under New Renewable Portfolio Standard, Peng Wang, Xiaohua Song, Haowen Yang, Xiaoying Zhai, Jingjing Han, Liwei Ju
Journal of System Simulation
Abstract: In order to clarify the multi-scale market coupling interaction relationship of electricity, green certificate, excess consumption under the renewable portfolio standards (RPS), the system dynamics is introduced, and the interactive trading model is constructed and simulated. Taking the logistics transformation of three market transaction targets as the clue, this paper designs a multi-scale market coupling transaction framework, constructs the complex causality of coupling transaction based on system dynamics method, and analyzes the impact of RPS on the revenue or cost of market participants. Simulation results show that under the new RPS, the electricity price will gradually decline, the …
Unoapi: Balancing Performance, Portability, And Productivity (P3) In Hpc Education, Konstantin Laufer, George K. Thiruvathukal
Unoapi: Balancing Performance, Portability, And Productivity (P3) In Hpc Education, Konstantin Laufer, George K. Thiruvathukal
Computer Science: Faculty Publications and Other Works
oneAPI is a major initiative by Intel aimed at making it easier to program heterogeneous architectures used in high-performance computing using a unified application programming interface (API). While raising the abstraction level via a unified API represents a promising step for the current generation of students and practitioners to embrace high- performance computing, we argue that a curriculum of well- developed software engineering methods and well-crafted exem- plars will be necessary to ensure interest by this audience and those who teach them. We aim to bridge the gap by developing a curriculum—codenamed UnoAPI—that takes a more holistic approach by looking …
Physics-Informed Neural Networks For Informed Vaccine Distribution In Heterogeneously Mixed Populations, Alvan Arulandu, Padmanabhan Seshaiyer
Physics-Informed Neural Networks For Informed Vaccine Distribution In Heterogeneously Mixed Populations, Alvan Arulandu, Padmanabhan Seshaiyer
Annual Symposium on Biomathematics and Ecology Education and Research
No abstract provided.
Continual Learning With Neural Networks, Pham Hong Quang
Continual Learning With Neural Networks, Pham Hong Quang
Dissertations and Theses Collection (Open Access)
Recent years have witnessed tremendous successes of artificial neural networks in many applications, ranging from visual perception to language understanding. However, such achievements have been mostly demonstrated on a large amount of labeled data that is static throughout learning. In contrast, real-world environments are always evolving, where new patterns emerge and the older ones become inactive before reappearing in the future. In this respect, continual learning aims to achieve a higher level of intelligence by learning online on a data stream of several tasks. As it turns out, neural networks are not equipped to learn continually: they lack the ability …
Daot: Domain-Agnostically Aligned Optimal Transport For Domain-Adaptive Crowd Counting, Huilin Zhu, Jingling Yuan, Xian Zhong, Zhengwei Yang, Zheng Wang, Shengfeng He
Daot: Domain-Agnostically Aligned Optimal Transport For Domain-Adaptive Crowd Counting, Huilin Zhu, Jingling Yuan, Xian Zhong, Zhengwei Yang, Zheng Wang, Shengfeng He
Research Collection School Of Computing and Information Systems
Domain adaptation is commonly employed in crowd counting to bridge the domain gaps between different datasets. However, existing domain adaptation methods tend to focus on inter-dataset differences while overlooking the intra-differences within the same dataset, leading to additional learning ambiguities. These domain-agnostic factors,e.g., density, surveillance perspective, and scale, can cause significant in-domain variations, and the misalignment of these factors across domains can lead to a drop in performance in cross-domain crowd counting. To address this issue, we propose a Domain-agnostically Aligned Optimal Transport (DAOT) strategy that aligns domain-agnostic factors between domains. The DAOT consists of three steps. First, individual-level differences …
Efficient Navigation For Constrained Shortest Path With Adaptive Expansion Control, Wenwen Xia, Yuchen Li, Wentian Guo, Shenghong Li
Efficient Navigation For Constrained Shortest Path With Adaptive Expansion Control, Wenwen Xia, Yuchen Li, Wentian Guo, Shenghong Li
Research Collection School Of Computing and Information Systems
In many route planning applications, finding constrained shortest paths (CSP) is an important and fundamental problem. CSP aims to find the shortest path between two nodes on a graph while satisfying a path constraint. Solving CSPs requires a large search space and is prohibitively slow on large graphs, even with the state-of-the-art parallel solution on GPUs. The reason lies in the lack of effective navigational information and pruning strategies in the search procedure. In this paper, we propose SPEC, a Shortest Path Enhanced approach for solving the exact CSP problem. Our design rationales of SPEC rely on the observation that …
Meta-Complementing The Semantics Of Short Texts In Neural Topic Models, Ce Zhang, Hady Wirawan Lauw
Meta-Complementing The Semantics Of Short Texts In Neural Topic Models, Ce Zhang, Hady Wirawan Lauw
Research Collection School Of Computing and Information Systems
Topic models infer latent topic distributions based on observed word co-occurrences in a text corpus. While typically a corpus contains documents of variable lengths, most previous topic models treat documents of different lengths uniformly, assuming that each document is sufficiently informative. However, shorter documents may have only a few word co-occurrences, resulting in inferior topic quality. Some other previous works assume that all documents are short, and leverage external auxiliary data, e.g., pretrained word embeddings and document connectivity. Orthogonal to existing works, we remedy this problem within the corpus itself by proposing a Meta-Complement Topic Model, which improves topic quality …
Graph Neural Network With Self-Attention And Multi-Task Learning For Credit Default Risk Prediction, Zihao Li, Xianzhi Wang, Lina Yao, Yakun Chen, Guandong Xu, Ee-Peng Lim
Graph Neural Network With Self-Attention And Multi-Task Learning For Credit Default Risk Prediction, Zihao Li, Xianzhi Wang, Lina Yao, Yakun Chen, Guandong Xu, Ee-Peng Lim
Research Collection School Of Computing and Information Systems
We propose a graph neural network with self-attention and multi-task learning (SaM-GNN) to leverage the advantages of deep learning for credit default risk prediction. Our approach incorporates two parallel tasks based on shared intermediate vectors for input vector reconstruction and credit default risk prediction, respectively. To better leverage supervised data, we use self-attention layers for feature representation of categorical and numeric data; we further link raw data into a graph and use a graph convolution module to aggregate similar information and cope with missing values during constructing intermediate vectors. Our method does not heavily rely on feature engineering work and …
A Quality Metric For K-Means Clustering Based On Centroid Locations, Manoj Thulasidas
A Quality Metric For K-Means Clustering Based On Centroid Locations, Manoj Thulasidas
Research Collection School Of Computing and Information Systems
K-Means clustering algorithm does not offer a clear methodology to determine the appropriate number of clusters; it does not have a built-in mechanism for K selection. In this paper, we present a new metric for clustering quality and describe its use for K selection. The proposed metric, based on the locations of the centroids, as well as the desired properties of the clusters, is developed in two stages. In the initial stage, we take into account the full covariance matrix of the clustering variables, thereby making it mathematically similar to a reduced chi2. We then extend it to account for …