Artificial Intelligence and Robotics Commons^™

Comparing Learned Representations Between Unpruned And Pruned Deep Convolutional Neural Networks, Parker Mitchell

Master's Theses

While deep neural networks have shown impressive performance in computer vision tasks, natural language processing, and other domains, the sizes and inference times of these models can often prevent them from being used on resource-constrained systems. Furthermore, as these networks grow larger in size and complexity, it can become even harder to understand the learned representations of the input data that these networks form through training. These issues of growing network size, increasing complexity and runtime, and ambiguity in the understanding of internal representations serve as guiding points for this work.

In this thesis, we create a neural network that …

Multi-Agent Pathfinding In Mixed Discrete-Continuous Time And Space, Thayne T. Walker

Electronic Theses and Dissertations

In the multi-agent pathfinding (MAPF) problem, agents must move from their current locations to their individual destinations while avoiding collisions. Ideally, agents move to their destinations as quickly and efficiently as possible. MAPF has many real-world applications such as navigation, warehouse automation, package delivery and games. Coordination of agents is necessary in order to avoid conflicts, however, it can be very computationally expensive to find mutually conflict-free paths for multiple agents – especially as the number of agents is increased. Existing state-ofthe- art algorithms have been focused on simplified problems on grids where agents have no shape or volume, and …

Energy Planning Model Design For Forecasting The Final Energy Consumption Using Artificial Neural Networks, Haidy Eissa

Theses and Dissertations

“Energy Trilemma” has recently received an increasing concern among policy makers. The trilemma conceptual framework is based on three main dimensions: environmental sustainability, energy equity, and energy security. Energy security reflects a nation’s capability to meet current and future energy demand. Rational energy planning is thus a fundamental aspect to articulate energy policies. The energy system is huge and complex, accordingly in order to guarantee the availability of energy supply, it is necessary to implement strategies on the consumption side. Energy modeling is a tool that helps policy makers and researchers understand the fluctuations in the energy system. Over the …

Prediction Of Iraqi Stock Exchange Using Optimized Based-Neural Network, Ameer Al-Haq Al-Shamery, Prof. Dr. Eman Salih Al-Shamery

Karbala International Journal of Modern Science

Stock market prediction is an interesting financial topic that has attracted the attention of researchers for the last years. This paper aims at improving the prediction of the Iraq-Stock-Exchange (ISX) using a developed method of feedforward Neural-Networks based on the Quasi-Newton optimization approach. The proposed method reduces the error factor depending on the Jacobian vector and Lagrange multiplier. This improvement has led to accelerating convergence during the learning process. A sample of companies listed on ISX was selected. This includes twenty-six banks for the years from 2010 to 2020. To evaluate the proposed model, the research findings are compared with …

Adadeep: A Usage-Driven, Automated Deep Model Compression Framework For Enabling Ubiquitous Intelligent Mobiles, Sicong Liu, Junzhao Du, Kaiming Nan, Zimu Zhou, Hui Liu, Zhangyang Wang, Yingyan Lin

Research Collection School Of Computing and Information Systems

Recent breakthroughs in deep neural networks (DNNs) have fueled a tremendously growing demand for bringing DNN-powered intelligence into mobile platforms. While the potential of deploying DNNs on resource-constrained platforms has been demonstrated by DNN compression techniques, the current practice suffers from two limitations: 1) merely stand-alone compression schemes are investigated even though each compression technique only suit for certain types of DNN layers; and 2) mostly compression techniques are optimized for DNNs’ inference accuracy, without explicitly considering other application-driven system performance (e.g., latency and energy cost) and the varying resource availability across platforms (e.g., storage and processing capability). To this …

Orthogonal Inductive Matrix Completion, Antoine Ledent, Rrodrigo Alves, Marius Kloft

Research Collection School Of Computing and Information Systems

We propose orthogonal inductive matrix completion (OMIC), an interpretable approach to matrix completion based on a sum of multiple orthonormal side information terms, together with nuclear-norm regularization. The approach allows us to inject prior knowledge about the singular vectors of the ground-truth matrix. We optimize the approach by a provably converging algorithm, which optimizes all components of the model simultaneously. We study the generalization capabilities of our method in both the distribution-free setting and in the case where the sampling distribution admits uniform marginals, yielding learning guarantees that improve with the quality of the injected knowledge in both cases. As …

Scheduling Allocation And Inventory Replenishment Problems Under Uncertainty: Applications In Managing Electric Vehicle And Drone Battery Swap Stations, Amin Asadi

Graduate Theses and Dissertations

In this dissertation, motivated by electric vehicle (EV) and drone application growth, we propose novel optimization problems and solution techniques for managing the operations at EV and drone battery swap stations. In Chapter 2, we introduce a novel class of stochastic scheduling allocation and inventory replenishment problems (SAIRP), which determines the recharging, discharging, and replacement decisions at a swap station over time to maximize the expected total profit. We use Markov Decision Process (MDP) to model SAIRPs facing uncertain demands, varying costs, and battery degradation. Considering battery degradation is crucial as it relaxes the assumption that charging/discharging batteries do not …

Yard Layout Optimization For General Cargo Terminal, Zhixiong Liu, Dong Yu, Chunjun Zhang

Journal of System Simulation

Abstract: Yard layout is an important component of the port yard allocation decision which affects the cargo storage capacity and through capacity for the port yard. As to the general cargo yard, combined with the cargo type and the yard storage strategy, the yard layout optimization model for the general cargo terminal is presented based on the statistical analysis for the production data when the optimization aim is minimizing the total horizontal transport distance of the trailer. The yard layout optimization results are employed by the mathematical tool Gurobi for different storage strategies, and the yard layout optimization results are …

Can Parallel Gravitational Search Algorithm Effectively Choose Parameters For Photovoltaic Cell Current Voltage Characteristics?, Alan Kirkpatrick

Honors Projects

This study asks the question “Can parallel Gravitational Search Algorithm (GSA) effectively choose parameters for photovoltaic cell current voltage characteristics?” These parameters will be plugged into the Single Diode Model to create the IV curve. It will also investigate Particle Swarm Optimization (PSO) and a population based random search (PBRS) to see if GSA performs the search better and or more quickly than alternative algorithms

Machine Learning Models For Deciphering Regulatory Mechanisms And Morphological Variations In Cancer, Saman Farahmand

Graduate Doctoral Dissertations

The exponential growth of multi-omics biological datasets is resulting in an emerging paradigm shift in fundamental biological research. In recent years, imaging and transcriptomics datasets are increasingly incorporated into biological studies, pushing biology further into the domain of data-intensive-sciences. New approaches and tools from statistics, computer science, and data engineering are profoundly influencing biological research. Harnessing this ever-growing deluge of multi-omics biological data requires the development of novel and creative computational approaches. In parallel, fundamental research in data sciences and Artificial Intelligence (AI) has advanced tremendously, allowing the scientific community to generate a massive amount of knowledge from data. Advances …

Adaptive Operating Hours For Improved Performance Of Taxi Fleets, Rajiv Ranjan Kumar, Pradeep Varakantham, Shih-Fen Cheng

Research Collection School Of Computing and Information Systems

Taxi fleets and car aggregation systems are an important component of the urban public transportation system. Taxis and cars in taxi fleets and car aggregation systems (e.g., Uber) are dependent on a large number of self-controlled and profit-driven taxi drivers, which introduces inefficiencies in the system. There are two ways in which taxi fleet performance can be optimized: (i) Operational decision making: improve assignment of taxis/cars to customers, while accounting for future demand; (ii) strategic decision making: optimize operating hours of (taxi and car) drivers. Existing research has primarily focused on the operational decisions in (i) and we focus on …

Learning To Fuse Asymmetric Feature Maps In Siamese Trackers, Wencheng Han, Xingping Dong, Fahad Shahbaz Khan, Ling Shao, Jianbing Shen

Computer Vision Faculty Publications

Recently, Siamese-based trackers have achieved promising performance in visual tracking. Most recent Siamese-based trackers typically employ a depth-wise cross-correlation (DW-XCorr) to obtain multi-channel correlation information from the two feature maps (target and search region). However, DW-XCorr has several limitations within Siamese-based tracking: it can easily be fooled by distractors, has fewer activated channels and provides weak discrimination of object boundaries. Further, DW-XCorr is a handcrafted parameter-free module and cannot fully benefit from offline learning on large-scale data. We propose a learnable module, called the asymmetric convolution (ACM), which learns to better capture the semantic correlation information in offline training on …

A Survey Of Edge Computing Resource Allocation And Task Scheduling Optimization, Wang Ling, Chuge Wu, Wenhui Fan

Journal of System Simulation

Abstract: With the rapid development of Internet of Things (IoT) and mobile terminals, the concept of edge computing arises. By moving the computation and storage capacity to the edge of network, edge computing is able to deal with a large amount of data produced by IoT devices and the responsive request from IoT application. To improve the utility of edge resource, the quality of service and quality of user experience, resource allocation and task scheduling optimization problems under edge computing attract wide attention. It becomes more difficult due to the geographic separated and heterogeneous features of edge computing resource as …

Molecule Optimization By Explainable Evolution, Binghong Chen, Tianzhe Wang, Chengtao Li, Hanjun Dai, Le Song

Machine Learning Faculty Publications

Optimizing molecules for desired properties is a fundamental yet challenging task in chemistry, material science, and drug discovery. This paper develops a novel algorithm for optimizing molecular properties via an Expectation-Maximization (EM) like explainable evolutionary process. The algorithm is designed to mimic human experts in the process of searching for desirable molecules and alternate between two stages: the first stage on explainable local search which identifies rationales, i.e., critical subgraph patterns accounting for desired molecular properties, and the second stage on molecule completion which explores the larger space of molecules containing good rationales. We test our approach against various baselines …

Simulation Research On Armored Equipment Maintenance Support Resource Optimization, Huiqi Zhang, Chunliang Chen, Junyan Liu, Liu Shuai, Yongqing Zhang

Journal of System Simulation

Abstract: Contraposing the problem of armored equipment maintenance resource computation and optimization, the armored equipment maintenance support process was analyzed. The armored equipment maintenance support process concept model and the mathematic model were constructed based on discrete system modeling theory; while the simulation model based on net discrete event was given in apply of Anylogic software. The armored equipment maintenance support task fulfilling degree, the resource utilization degree, and the mean maintenance time were got by simulation computation. As a result that maintenance support resource was evaluated based on index weight, and the optimized values of the maintenance support resource …

Quadratic Rational Trigonometric Spline Curves With Shape Controlling, Xinru Liu, Manman Wei, Shengjun Liu, Dangfu Yang

Journal of System Simulation

Abstract: A new quadratic rational trigonometric spline curve with a shape parameter was proposed. The value control and the inflection-point control of the interpolation scheme were discussed in theory. And the optimal methods for calculating the desired inflection-points was proposed, by using optimization theory. Numerical experiments show the interpolation spline and the optimization method can be used in modeling design.

Adaptive Quick Artificial Bee Colony Algorithm Based On Opposition Learning, Xiaojian Yang, Yiwei Dong

Journal of System Simulation

Abstract: On the basis of analyzing such shortcomings of the artificial bee colony algorithm (ABC) as slow convergence, low convergence precision and premature convergence, the opposition-learning adaptive quick artificial bee colony algorithm (OAQABC) was proposed. A new step size was proposed, which made the around food source parameter of quick artificial bee colony algorithm (QABC) adaptive, and combined the opposition-based learning to improve the employed bee phase. The experimental results show that OAQABC has better performance than basic ABC and QABC. Also the optimization performance of OAQABC is better than particle swarm optimization (PSO) algorithm and Cuckoo Search (CS) algorithm …

Affinity Propagation Based Improved Group Search Optimizer Clustering Algorithm, Zhang Kang, Xingsheng Gu

Journal of System Simulation

Abstract: The essence of clustering is an optimization problem. It can be solved by swarm intelligent algorithms which are the popular research area in recent years. A novel Group Search Optimizer (GSO) algorithm named Fast Global Group Search Optimizer (FGGSO) was proposed. FGGSO improved the individuals' updating strategies of GSO, adopting the campaign strategy, destruction-construction strategy and accelerating-jumping strategy. By this means, the proposed algorithm improved the global and local search capability of the original GSO. Furthermore, based on this FGGSO algorithm, a novel improved AP algorithm was proposed. On account of deficiency of AP clustering unable to deal with …

Least-Energy Maneuver Of Five-Link Manipulator Constrained Within Tunnel Space Using Direct Collocation, Xiuqiang Pan, Chengcai Mei, Junjie Chen

Journal of System Simulation

Abstract: Optimal control and designs least-energy maneuver control laws for a five-linked manipulator were applied in order to carry out designated tasks in a confined space. Lagrange-Euler equation described the relationships between the actuators and system dynamics. Euler-Lagrange formulation indicates how optimization can be achieved when optimum occurs. Direct collocation method was introduced in order to solve this highly nonlinear dynamic optimal control problem. Simulations were done to exploit how the manipulator reacted to the constraint. In this study, the diameter of the cylindrical space was shrunken each time by 0.1 meters. The value of the cost function and …

Optimization Scheme Of Average Time For Finding Idle Channel In Cognitive Radio System, Qiao Pei, Liyuan Xiao, Yanyan Han, Gao Ling

Journal of System Simulation

Abstract: In cognitive radio system, periodic spectrum sensing was taken by secondary users to prevent the interference to primary users. Supposed that there are many primary user channels, when the current primary user occupies channel, secondary users do spectrum handover. During spectrum handover, the time of finding an idle channel is a random variable. In order to speed up spectrum handover, the system used equal gain combining cooperative spectrum sensing to inspect an idle channel. This pattern optimized the sensing time of the single user channel, in order to get the best effect of the average time in finding idle …

Parameters Optimization For Variable Speed And Pitch Controller Of Wind Turbine Based On Bladed, Gao Feng, Wang Wei, Xinmei Ling

Journal of System Simulation

Abstract: Due to nonlinearity and time-varying parameters of wind power system, its controller parameters are hard to be calculated and tuned during the process of design and optimization. The linear model which is suitable for parameters tuning was built through model linearization of Bladed and model reducing-order algorithm. The PI parameter was tuned with the IM-PSO (Immune Memory Particle Swarm Optimization). Moreover, the gain coefficient of optimal torque control and the gain divisor of adaptive PI pitch control conducted optimizing calculation based on the identification parameters of Bladed. A set of optimization method for variable speed and pitch controller …

Scheduling Problem Of Unidirectional Material Handling System With Short-Cut, Juntao Li, Kun Xia, Kise Hiroshi

Journal of System Simulation

Abstract: Unidirectional circulation-type material handling system on a single loop with a shortcut is a typical and basic unit in a flexible manufacturing system or logistics system. It is widely used in semiconductor wafer fabrication system. Superposition efficiency of the basic unit often determines the one of the whole system. Different scheduling rules impact the interference between AGVs and then have an important effect on the efficiency of the whole system. To decrease the interference and improve performance of the system, a mathematical model of this system was made, analyzing this system by simulation and comparing the merits and …

Real-Time Control Method Of Hsss Based On Single Phrase Self Decoupling Strategy, Yingping Yi, Bogang Qu, Zhang Yang

Journal of System Simulation

Abstract: Zero Crossing Detection (ZCD) and phase locked loop are widely applied in the real-time control. Compared the performances of ZCD, SSRF SPLL and DDSRF SPLL,a method named as Single Phase Self Decoupling SPLL (SPSD SPLL) for HSSS(Hybrid Solid State Switch) is proposed and the mathematical method and control methods are given. The control parameters were obtained by analyzing the steady and dynamic performance of SPSD SPLL. By establishing the real-time controlling models based on the above strategies and HSSS model in the MATLAB/Simulink, the simulation and optimization analysis results verified that the SPSD SPLL was well performed …

Study On Vacuum Dehydration Rate From Oil Based On T_S Fuzzy Identifying Model, Liu Ge, Bin Chen, Xianming Zhang

Journal of System Simulation

Abstract: The process of vacuum dehydration from oil is time-varying, nonlinear, and difficult to be specified with mathematical methods. Takagi-Sugeno (T_S) fuzzy model of vacuum dehydration rate of oil purifier is proposed, which a method of applying Fuzzy C-Means (FCM) clustering algorithm and using the least square method identifying the consequent parameters. The nonlinear mapping is set up from four influence factors (the initial water content, the vacuum pressure , the initial temperature and running time) to vacuum dehydration rate using the T_S fuzzy model. The simulation and experimental results show the T_S model reflects the laws of the influences …

Improved Particle Swarm Optimization Based On Lévy Flights, Rongyu Li, Wang Ying

Journal of System Simulation

Abstract: The particle swarm optimization (PSO) has some demerits, such as relapsing into local extremum, slow convergence velocity and low convergence precision in the late evolutionary. The Lévy particle swarm optimization (Lévy PSO) was proposed. In the particle position updating formula, Lévy PSO eliminated the impact of speed on the convergence rate, and used Levy flight to change the direction of particle positions movement to prevent particles getting into local optimum value, and then using greedy strategy to update the evaluation and choose the best solution to obtain the global optimum. The experimental results show that Lévy PSO can effectively …

Computational Model For Neural Architecture Search, Ram Deepak Gottapu

Doctoral Dissertations

"A long-standing goal in Deep Learning (DL) research is to design efficient architectures for a given dataset that are both accurate and computationally inexpensive. At present, designing deep learning architectures for a real-world application requires both human expertise and considerable effort as they are either handcrafted by careful experimentation or modified from a handful of existing models. This method is inefficient as the process of architecture design is highly time-consuming and computationally expensive.

The research presents an approach to automate the process of deep learning architecture design through a modeling procedure. In particular, it first introduces a framework that treats …

Catgame: A Tool For Problem Solving In Complex Dynamic Systems Using Game Theoretic Knowledge Distribution In Cultural Algorithms, And Its Application (Catneuro) To The Deep Learning Of Game Controller, Faisal Waris

Wayne State University Dissertations

Cultural Algorithms (CA) are knowledge-intensive, population-based stochastic optimization methods that are modeled after human cultures and are suited to solving problems in complex environments. The CA Belief Space stores knowledge harvested from prior generations and re-distributes it to future generations via a knowledge distribution (KD) mechanism. Each of the population individuals is then guided through the search space via the associated knowledge. Previously, CA implementations have used only competitive KD mechanisms that have performed well for problems embedded in static environments. Relatively recently, CA research has evolved to encompass dynamic problem environments. Given increasing environmental complexity, a natural question arises …

Credible Optimum Selection Of Guidance System Simulation Based On Entropy Weight Vikor Method, Wenguang Yang, Yunjie Wu

Journal of System Simulation

Abstract: As the core component of the missile system, the guidance system plays an increasingly important role in the design of missile system. In order to improve the accuracy of the guidance system, this paper obtains the experimental data under a number of parameter design schemes by means of simulation experiments. The problem of simulation credibility verification of guidance system is transformed into multi-attribute decision-making optimization problem, and a parameter optimization method of guidance system based on improved VIKOR method is designed. The improved VIKOR method overcomes the phenomenon of rank reversal and ensures that the optimal final compromise solution …

Optimal Sampling Paths For Autonomous Vehicles In Uncertain Ocean Flows, Andrew J. De Stefan

Dissertations

Despite an extensive history of oceanic observation, researchers have only begun to build a complete picture of oceanic currents. Sparsity of instrumentation has created the need to maximize the information extracted from every source of data in building this picture. Within the last few decades, autonomous vehicles, or AVs, have been employed as tools to aid in this research initiative. Unmanned and self-propelled, AVs are capable of spending weeks, if not months, exploring and monitoring the oceans. However, the quality of data acquired by these vehicles is highly dependent on the paths along which they collect their observational data. The …

Re-Org: An Online Repositioning Guidance Agent, Muralidhar Konda, Pradeep Varakantham, Aayush Saxena, Meghna Lowalekar

Research Collection School Of Computing and Information Systems

No abstract provided.