Artificial Intelligence and Robotics Commons^™

A Mixed-Integer Linear Programming Reduction Of Disjoint Bilinear Programs Via Symbolic Variable Elimination, Jihwan Jeong, Scott Sanner, Akshat Kumar

Research Collection School Of Computing and Information Systems

A disjointly constrained bilinear program (DBLP) has various practical and industrial applications, e.g., in game theory, facility location, supply chain management, and multi-agent planning problems. Although earlier work has noted the equivalence of DBLP and mixed-integer linear programming (MILP) from an abstract theoretical perspective, a practical and exact closed-form reduction of a DBLP to a MILP has remained elusive. Such explicit reduction would allow us to leverage modern MILP solvers and techniques along with their solution optimality and anytime approximation guarantees. To this end, we provide the first constructive closed-form MILP reduction of a DBLP by extending the technique of …

The Bemi Stardust: A Structured Ensemble Of Binarized Neural Networks, Ambrogio Maria Bernardelli, Stefano Gualandi, Hoong Chuin Lau, Simone Milanesi

Research Collection School Of Computing and Information Systems

Binarized Neural Networks (BNNs) are receiving increasing attention due to their lightweight architecture and ability to run on low-power devices, given the fact that they can be implemented using Boolean operations. The state-of-the-art for training classification BNNs restricted to few-shot learning is based on a Mixed Integer Programming (MIP) approach. This paper proposes the BeMi ensemble, a structured architecture of classification-designed BNNs based on training a single BNN for each possible pair of classes and applying a majority voting scheme to predict the final output. The training of a single BNN discriminating between two classes is achieved by a MIP …

Imitating Opponent To Win: Adversarial Policy Imitation Learning In Two-Player Competitive Games, The Viet Bui, Tien Mai, Thanh H. Nguyen

Research Collection School Of Computing and Information Systems

Recent research on vulnerabilities of deep reinforcement learning (RL) has shown that adversarial policies adopted by an adversary agent can influence a target RL agent (victim agent) to perform poorly in a multi-agent environment. In existing studies, adversarial policies are directly trained based on experiences of interacting with the victim agent. There is a key shortcoming of this approach --- knowledge derived from historical interactions may not be properly generalized to unexplored policy regions of the victim agent, making the trained adversarial policy significantly less effective. In this work, we design a new effective adversarial policy learning algorithm that overcomes …

Curricular Contrastive Regularization For Physics-Aware Single Image Dehazing, Yu Zheng, Jiahui Zhan, Shengfeng He, Yong Du

Research Collection School Of Computing and Information Systems

Considering the ill-posed nature, contrastive regularization has been developed for single image dehazing, introducing the information from negative images as a lower bound. However, the contrastive samples are non-consensual, as the negatives are usually represented distantly from the clear (i.e., positive) image, leaving the solution space still under-constricted. Moreover, the interpretability of deep dehazing models is underexplored towards the physics of the hazing process. In this paper, we propose a novel curricular contrastive regularization targeted at a consensual contrastive space as opposed to a non-consensual one. Our negatives, which provide better lower-bound constraints, can be assembled from 1) the hazy …

Where Is My Spot? Few-Shot Image Generation Via Latent Subspace Optimization, Chenxi Zheng, Bangzhen Liu, Huaidong Zhang, Xuemiao Xu, Shengfeng He

Research Collection School Of Computing and Information Systems

Image generation relies on massive training data that can hardly produce diverse images of an unseen category according to a few examples. In this paper, we address this dilemma by projecting sparse few-shot samples into a continuous latent space that can potentially generate infinite unseen samples. The rationale behind is that we aim to locate a centroid latent position in a conditional StyleGAN, where the corresponding output image on that centroid can maximize the similarity with the given samples. Although the given samples are unseen for the conditional StyleGAN, we assume the neighboring latent subspace around the centroid belongs to …

Venus: A Geometrical Representation For Quantum State Visualization, Shaolun Ruan, Ribo Yuan, Qiang Guan, Yanna Lin, Ying Mao, Weiwen Jiang, Zhepeng Wang, Wei Xu, Yong Wang

Research Collection School Of Computing and Information Systems

Visualizations have played a crucial role in helping quantum computing users explore quantum states in various quantum computing applications. Among them, Bloch Sphere is the widely-used visualization for showing quantum states, which leverages angles to represent quantum amplitudes. However, it cannot support the visualization of quantum entanglement and superposition, the two essential properties of quantum computing. To address this issue, we propose VENUS, a novel visualization for quantum state representation. By explicitly correlating 2D geometric shapes based on the math foundation of quantum computing characteristics, VENUS effectively represents quantum amplitudes of both the single qubit and two qubits for quantum …

Tracing The Twenty-Year Evolution Of Developing Ai For Eye Screening In Singapore: A Master Chronology Of Sidrp, Selena+ And Eyris, Steven M. Miller

Research Collection School Of Computing and Information Systems

This working paper is entirely comprised of a timeline table that begins in 2002 and runs through mid-2023. Across these two decades, this timeline traces the evolutionary development of the following:

• The early Singapore R&D efforts to apply software-based image analysis algorithms and methods to analyse eye retina images for diabetic retinopathy and other eye diseases. This was based on a collaboration between the Singapore Eye Research Institute (SERI) and its parent organization, the Singapore National Eye Centre (SNEC), with faculty from the School of Computing at National University of Singapore.
• The establishment and operation of the Singapore Integrated Diabetic …

Mosaic: Spatially-Multiplexed Edge Ai Optimization Over Multiple Concurrent Video Sensing Streams, Ila Gokarn, Hemanth Sabbella, Yigong Hu, Tarek Abdelzaher, Archan Misra

Research Collection School Of Computing and Information Systems

Sustaining high fidelity and high throughput of perception tasks over vision sensor streams on edge devices remains a formidable challenge, especially given the continuing increase in image sizes (e.g., generated by 4K cameras) and complexity of DNN models. One promising approach involves criticality-aware processing, where the computation is directed selectively to "critical" portions of individual image frames. We introduce MOSAIC, a novel system for such criticality-aware concurrent processing of multiple vision sensing streams that provides a multiplicative increase in the achievable throughput with negligible loss in perception fidelity. MOSAIC determines critical regions from images received from multiple vision …

Machine Learning And Network Embedding Methods For Gene Co-Expression Networks, Niloofar Aghaieabiane

Dissertations

High-throughput technologies such as DNA microarrays and RNA-seq are used to measure the expression levels of large numbers of genes simultaneously. To support the extraction of biological knowledge, individual gene expression levels are transformed into Gene Co-expression Networks (GCNs). GCNs are analyzed to discover gene modules. GCN construction and analysis is a well-studied topic, for nearly two decades. While new types of sequencing and the corresponding data are now available, the software package WGCNA and its most recent variants are still widely used, contributing to biological discovery.

The discovery of biologically significant modules of genes from raw expression data is …

Deep Hybrid Modeling Of Neuronal Dynamics Using Generative Adversarial Networks, Soheil Saghafi

Dissertations

Mechanistic modeling and machine learning methods are powerful techniques for approximating biological systems and making accurate predictions from data. However, when used in isolation these approaches suffer from distinct shortcomings: model and parameter uncertainty limit mechanistic modeling, whereas machine learning methods disregard the underlying biophysical mechanisms. This dissertation constructs Deep Hybrid Models that address these shortcomings by combining deep learning with mechanistic modeling. In particular, this dissertation uses Generative Adversarial Networks (GANs) to provide an inverse mapping of data to mechanistic models and identifies the distributions of mechanistic model parameters coherent to the data.

Chapter 1 provides background information on …

What Effects Do Large Language Models Have On Cybersecurity, Josiah Marshall

Cybersecurity Undergraduate Research Showcase

Large Language Models (LLMs) are artificial intelligence (AI) tools that can process, summarize, and translate texts and predict future words in a sentence, letting the LLM generate sentences similar to how humans talk and write. One concern that needs to be flagged is that, often, the content generated by different LLMs is inaccurate. LLMs are trained on code that can be used to detect data breaches, detect ransomware, and even pinpoint organizational vulnerabilities in advance of a cyberattack. LLMs are new but have unbelievable potential with their ability to generate code that brings awareness to cyber analysts and IT professionals. …

Data-Optimized Spatial Field Predictions For Robotic Adaptive Sampling: A Gaussian Process Approach, Zachary Nathan

Computer Science Senior Theses

We introduce a framework that combines Gaussian Process models, robotic sensor measurements, and sampling data to predict spatial fields. In this context, a spatial field refers to the distribution of a variable throughout a specific area, such as temperature or pH variations over the surface of a lake. Whereas existing methods tend to analyze only the particular field(s) of interest, our approach optimizes predictions through the effective use of all available data. We validated our framework on several datasets, showing that errors can decline by up to two-thirds through the inclusion of additional colocated measurements. In support of adaptive sampling, …

Rapid Assessment Of Fish Freshness For Multiple Supply-Chain Nodes Using Multi-Mode Spectroscopy And Fusion-Based Artificial Intelligence, Hossein Kashani Zadeh, Mike Hardy, Mitchell Sueker, Yicong Li, Angelis Tzouchas, Nicholas Mackinnon, Gregory Bearman, Simon A Haughey, Alireza Akhbardeh, Insuck Baek, Chansong Hwang, Jianwei Qin, Amanda M Tabb, Rosalee S Hellberg, Shereen Ismail, Hassan Reza, Fartash Vasefi, Moon Kim, Kouhyar Tavakolian, Christopher T Elliott

Faculty, Staff and Student Publications

This study is directed towards developing a fast, non-destructive, and easy-to-use handheld multimode spectroscopic system for fish quality assessment. We apply data fusion of visible near infra-red (VIS-NIR) and short wave infra-red (SWIR) reflectance and fluorescence (FL) spectroscopy data features to classify fish from fresh to spoiled condition. Farmed Atlantic and wild coho and chinook salmon and sablefish fillets were measured. Three hundred measurement points on each of four fillets were taken every two days over 14 days for a total of 8400 measurements for each spectral mode. Multiple machine learning techniques including principal component analysis, self-organized maps, linear and …

"Church On My Couch": Predicting The Future Impact Of Online Ministry Based On The Impact During Covid-19, Samukeliso Mabarani, Sikhumbuzo Dube

Adventist Human-Subject Researchers Association

With “everything from home” as the new norm, “how does the use of digital platforms impact Adventist education, community engagement, and spiritual outreach?” Using a quantitative approach, we draw insights from online ministry during Covid-19 and use the insights to predict the future impact of online ministry statistically.

Handling Realistic Label Noise In Bert Text Classification, Maha Tufail Agro, Hanan Al Darmaki

Natural Language Processing Faculty Publications

Label noise refers to errors in training labels caused by cheap data annotation methods, such as web scraping or crowd-sourcing, which can be detrimental to the performance of supervised classifiers. Several methods have been proposed to counteract the effect of random label noise in supervised classification, and some studies have shown that BERT is already robust against high rates of randomly injected label noise. However, real label noise is not random; rather, it is often correlated with input features or other annotator-specific factors. In this paper, we evaluate BERT in the presence of two types of realistic label noise: feature-dependent …

Simulation Of Real-Time Path Planning And Formation Control For Unmanned Surface Vessel, Dalei Song, Wenhao Gan, Yingzhi Xu, Xiuqing Qu, Jiangli Cao

Journal of System Simulation

Abstract: Safety and collision-free navigation are the basis of normal navigation of an unmanned surface vessel. The high-fidelity virtual ocean is constructed by using Unity3D.On the basis of the vessel modeling, a real-time path planning and formation control method for unknown complex environments is proposed. Firstly, the local environment information is obtained by the laser sensor. Then the real-time local path planning is completed by combining A-star and route-thinning methods under the replanning strategy. In addition, formation control is carried out based on the leader-follower strategy and consistency method, and the artificial potential field …

Research On Optimal Scheduling Of Microgrid Based On Nbbo Algorithm, Lisheng Wei, Benben Yang, Ruixia Sun

Journal of System Simulation

Abstract: In view of the economic and environmental collaborative optimization of a microgrid with a micro gas turbine, power to gas(P2G) system with the ability to abandon wind and light for consumption and capture carbon is introduced, and a microgrid optimal scheduling model with P2G system based on novel biogeography-based optimization(NBBO) algorithm is proposed. The microgrid model with the P2G system is constructed, and the working principle of the main equipment is analyzed. The reserve capacity of a wind turbine is introduced to reduce the influence of the randomness of wind power …

Deep Reinforcement Learning-Based Control Strategy For Boost Converter, Yuxuan Dai, Chenggang Cui

Journal of System Simulation

Abstract: In view of the problems of model uncertainty and nonlinearity in bus voltage stability control of Boost converter, an intelligent control strategy based on model-free deep reinforcement learning(RL) is proposed. RL double DQN(DDQN) algorithm and deep deterministic policy gradient(DDPG) algorithm are used, and the Boost converter controller is designed. The state, action space, reward function, and neural network are also designed to improve the dynamic performance of the controller. The joint simulation of the Boost converter model and RL agent is realized by RL modelica(RLM …

Point Cloud Registration Method Based On Improved Covariance Matrix Descriptor, Yuan Zhang, Haoyu Han, Xie Han, Jiaxu Fu

Journal of System Simulation

Abstract: Point cloud registration is a key part of the digital protection of cultural relics. Improving registration accuracy and noise resistance is the main goal of point cloud registration for cultural relics. In order to solve this problem, a three-dimensional (3D) point cloud registration method based on a covariance matrix descriptor is proposed. The tensor voting method is used to eliminate the noise points, and the internal shape signature method is used to extract the key points from the point cloud after removing the noise. Then, the neighborhood information is constructed for the extracted key points, …

Outlier Detection During Thermal Processes Based On Improved Gaussian Mixture Model, Zheng Wu, Yue Zhang, Ze Dong

Journal of System Simulation

Abstract: Abnormal data detection during thermal processes is the basis for performing system modeling, control, and optimization and constitutes an important part of data processing. In this paper, an unsupervised outlier detection algorithm during thermal processes based on an improved Gaussian mixture model is proposed. The algorithm captures a class of data clusters under specific working conditions by using Gaussian components in each dimension, modifies the posterior probability density of the traditional model by adding penalty constraint factors to penalize the false detection and missed detection items, and identifies abnormal data according to the correlation differences with the …

Improved Social Force Model Based On Enhancing Psych Behavioral Heterogeneity, Yandong Liu, Gaoxiang Huang, Wen Chen

Journal of System Simulation

Abstract: Simulating the evacuation behavior of people under anxiety is of great significance for solving the kinematic problems such as escape. At present, most at home and abroad studies consider the anxiety factors as the only medium of population evacuation without considering how external key factors affect anxiety factors in such emergency environments. The improved social force model is proposed, combined with Agent-based stampede risk assessment, the influence of key environmental variables on the anxiety factor is quantified. The psychological force parameters are introduced, and the impact of the anxiety factor on the actual evacuation process is applied to the …

Simulation And Optimization Of Integrated Production Logistics System Of Underground Coal Mining, Dressing, And Backfilling, Xiangqian Wang, Puhao Guo, Xiangrui Meng

Journal of System Simulation

Abstract: In order to study the green mining mode of coal, the bottleneck problem in the integrated production logistics system of underground coal mining, dressing, and backfilling under the goal of achieving the basic production capacity target is explored, so as to promote the coordinated and efficient logistics transportation of underground coal and gangue. A mine in Shanxi is selected as the prototype, and the queuing theory is adopted to analyze the operation process of the integrated coal production logistics system. A discrete event model is established with the help of Anylogic simulation software for related experimental optimization …

Research On Collaborative Task Allocation Method Of Multiple Uavs Based On Blockchain, Shuangcheng Niu, Yuqiang Jin, Kunhu Kou

Journal of System Simulation

Abstract: The autonomous collaborative control of a multi-unmanned aerial vehicle (UAV) system lacks a unified underlying technology platform and faces single point failure and information security threats. In order to solve these problems, an idea to build collaborative task planning platforms based on blockchain technology is proposed. With thecollaborative task allocation of multiple UAVs as research objects, an online, safe, high-efficiency, and real-time task allocation method is designed. The contract network task allocation algorithm is described as a smart contract, and system consensus is reached based on the blockchain consensus algorithm. In addition, …

Auxiliary Decision-Making Method For Flight Support Force Allocation Based On Dbn, Junyang Liu, Shisong Zhu

Journal of System Simulation

Abstract: In order to make up for the deficiency of decision-making methods of full deployment and proportional allocation in airport flight support activities, an auxiliary decision-making model for airport flight support force allocation based on a dynamic Bayesian network (DBN) is constructed.On the basis of quantifying the strength of the support force, thehidden Markov model and Hausdorff distance algorithm are introduced to make auxiliary decisions for flight support activities under different conditions and define relevant decision evaluation indexes as the basis for model verification. Simulation tests are carried out on airport flight support activities under two …

Two-Stage Distributed Robust Optimal Dispatching For A Combined Heat And Power Virtual Power Plant, Yaqian Fan, Songyuan Yu, Fang Fang

Journal of System Simulation

Abstract: Combined heat and power virtual power plant (CHP-VPP) aggregates various electrical and thermal output units and takes into account the uncertainty of wind and solar output, dynamic electricity prices, thermal comfort of users, and other influences to achieve optimal dispatching of overall output. A two-stage distributed robust optimal dispatching method is proposed. In the first stage, planned dispatching is considered, so as to maximize the benefit of CHP-VPP. In the second stage, a fuzzy set of wind and solar output uncertainties is constructed based on the distributed robust method of moment uncertainty, and …

Multi-Camera Vehicle Recognition Method Based On Feature Robustness Enhancement, Huicheng Luo, Shujuan Wang

Journal of System Simulation

Abstract: Due to factors such as viewpoint changes, complex environments, and pose differences under multiple cameras, the images of the same vehicle in different scenes show huge appearance ambiguity, which brings challenges to vehicle identity matching. In order to solve this problem, a feature robustness enhancement method for vehicle recognition is proposed under the transformer framework. Based on the fact that the structural information of the vehicle is invariant under multiple cameras, a module for enhancing structural information guided by contour features is designed, and a structural feature perception loss is proposed to promote the fusion …

Picking Path Planning Of Container Robots Based On Improved Genetic Algorithm, Yuwen Wu, Zhiyue Niu, Zhenping Li

Journal of System Simulation

Abstract: Under the new "container-to-person" picking mode in intelligent warehouses, a new optimization model and its improved genetic algorithm are proposed to solve the picking path planning problem of multiple container robots. According to the picking mode and characteristics of container robots, the picking path planning problem is transformed into an asymmetric vehicle routing problem, and a mixed integer programming model is established with bi-objectives of the shortest total picking path and the least completion time. A hybrid genetic algorithm is designed to solve this model, and the effectiveness and stability of the algorithm are verified …

Simulation Platform For Optimization And Decision Making Of Flexible Manufacturing Process Of Automatic Production Lines, Chao Fu, Dongyue Wang, Xinxin Li, Min Xue

Journal of System Simulation

Abstract: In order to simulate a flexible manufacturing process to verify the correctness of theoretical computation and analysis and find other important factors for optimization and decision making in the manufacturing process, a simulation platform for the optimization and decision making of the flexible manufacturing process of automatic production lines is designed and developed. The framework of the simulation platform is constructed with four hierarchies, including the enterprise production hierarchy, optimization and decision making hierarchy, data resource hierarchy, and function module hierarchy. On this basis, the functional implementation of the modules of production allocation …

Modeling And Simulation Of Spaceborne, Near-Spaceborne, And Airborne Integrated Collaborative Remote Sensing System Based On Dodaf, Lili An, Tian Xia, Wenbin Yang, Xinbo Wu

Journal of System Simulation

Abstract: Spaceborne, near-spaceborne, and airborne integrated collaborative remote sensing system (SNA-ICRSS) makes comprehensive use of modern information technology to aggregate multiple and heterogeneous data in spaceborne, near-spaceborne, and airborne domains, so as to realize accurate emergency service and command decision-making. SNA-ICRSS is huge and complex, and there is a lack of research on its architecture modeling and simulation. According to the architecture characteristics of the SNA-ICRSS, the minimum prototype of the SNA-ICRSS is constructed; through the department of defense architecture framework (DoDAF) and activity based methodology (ABM) methods, the minimal prototype of the SNA-ICRSS is modeled, and the operational resource …

An Algorithm For Obtaining Interception Guidance Routes Weighted By Threat Indexes, Shuyuan Liu

Journal of System Simulation

Abstract: Interception guidance and threat assessment are two inseparable parts of command and control. The former implements specific command and guidance through guidance solutions, and the latter provides basis for threat assessment and situation advantage to command and control. Taking threat indexes as the weight, a kind of optimal interception route index is proposed to obtain the guidance solution with the least threat. Simulation results show that the proposed algorithm can effectively provide routing solutions for interception with the least threat, and is reliable under different situations, or with different weights of threat indexes. The algorithm provides a new idea …