Artificial Intelligence and Robotics Commons^™

Learning And Evaluating Chinese Idiom Embeddings, Minghuan Tan, Jing Jiang

Research Collection School Of Computing and Information Systems

We study the task of learning and evaluating Chinese idiom embeddings. We first construct a new evaluation dataset that contains idiom synonyms and antonyms. Observing that existing Chinese word embedding methods may not be suitable for learning idiom embeddings, we further present a BERT-based method that directly learns embedding vectors for individual idioms. We empirically compare representative existing methods and our method. We find that our method substantially outperforms existing methods on the evaluation dataset we have constructed.

Dynamic Heterogeneous Graph Embedding Via Heterogeneous Hawkes Process, Yugang Ji, Tianrui Jia, Yuan Fang, Chuan Shi

Research Collection School Of Computing and Information Systems

Graph embedding, aiming to learn low-dimensional representations of nodes while preserving valuable structure information, has played a key role in graph analysis and inference. However, most existing methods deal with static homogeneous topologies, while graphs in real-world scenarios are gradually generated with different-typed temporal events, containing abundant semantics and dynamics. Limited work has been done for embedding dynamic heterogeneous graphs since it is very challenging to model the complete formation process of heterogeneous events. In this paper, we propose a novel Heterogeneous Hawkes Process based dynamic Graph Embedding (HPGE) to handle this problem. HPGE effectively integrates the Hawkes process into …

Artificial Intelligence And Work: Two Perspectives, Steven Miller, Thomas H. Davenport

Research Collection School Of Computing and Information Systems

One of the most important issues in contemporary societies is the impact of intelligent technologies on human work. For an empirical perspective on the issue, we recently completed 30 case studies of people collaborating with AI-enabled smart machines. Twenty-four were from North America, mostly in the US. Six were from Southeast Asia, mostly in Singapore. We compare some of our observations to one of the broadest academic examinations of the issue. In particular, we focus on our case study observations with regard to key findings from the MIT Task Force on the Work of the Future report.

Learning Environment Containerization Of Machine Learning For Cybersecurity, Hao Zhang

Symposium of Student Scholars

Machine learning plays a critical role in detecting and preventing in the field of cybersecurity. However, many students have difficulties on configuring the appropriate coding environment and retrieving datasets on their own computers, which, to some extent, wastes valuable time for learning core contents of machine learning and cybersecurity. In this paper, we propose an approach with learning environment containerization of machine learning algorithm and dataset. This will help students focus more on learning contents and have valuable hand-on experience through Docker container and get rid of the trouble of configuration coding environment and retrieve dataset. This paper provides an …

Spam Email Detection: Comparison Between Naïve Bayes And Neural Network, Zhuolin Li

Symposium of Student Scholars

Classification is an important technique to deal with cybersecurity threats. In this paper, we detect spam emails from publicly available dataset using Naive Bayes and Neural Network (NN). The results from experiments show that for data sets with more balanced for classification, the accuracy of Naive Bayes is better than NN

Gradient Free Sign Activation Zero One Loss Neural Networks For Adversarially Robust Classification, Yunzhe Xue

Dissertations

The zero-one loss function is less sensitive to outliers than convex surrogate losses such as hinge and cross-entropy. However, as a non-convex function, it has a large number of local minima, andits undifferentiable attribute makes it impossible to use backpropagation, a method widely used in training current state-of-the-art neural networks. When zero-one loss is applied to deep neural networks, the entire training process becomes challenging. On the other hand, a massive non-unique solution probably also brings different decision boundaries when optimizing zero-one loss, making it possible to fight against transferable adversarial examples, which is a common weakness in deep learning …

Towards Adversarial Robustness With 01 Lossmodels, And Novel Convolutional Neural Netsystems For Ultrasound Images, Meiyan Xie

Dissertations

This dissertation investigates adversarial robustness with 01 loss models and a novel convolutional neural net systems for vascular ultrasound images.

In the first part, the dissertation presents stochastic coordinate descent for 01 loss and its sensitivity to adversarial attacks. The study here suggests that 01 loss may be more resilient to adversarial attacks than the hinge loss and further work is required.

In the second part, this dissertation proposes sign activation network with a novel gradient-free stochastic coordinate descent algorithm and its ensembling model. The study here finds that the ensembling model gives a high minimum distortion (as measured by …

Data-Driven Learning For Robot Physical Intelligence, Leidi Zhao

Dissertations

The physical intelligence, which emphasizes physical capabilities such as dexterous manipulation and dynamic mobility, is essential for robots to physically coexist with humans. Much research on robot physical intelligence has achieved success on hyper robot motor capabilities, but mostly through heavily case-specific engineering. Meanwhile, in terms of robot acquiring skills in a ubiquitous manner, robot learning from human demonstration (LfD) has achieved great progress, but still has limitations handling dynamic skills and compound actions. In this dissertation, a composite learning scheme which goes beyond LfD and integrates robot learning from human definition, demonstration, and evaluation is proposed. This method tackles …

Advances In Deep Learning With Applications To Computer Vision And Astronomy, Zhihang Hu

Dissertations

Deep Learning has spanned a variety of applications in computer vision as well as computational astronomy. These two aspects obtained similar data structure, therefore, their solutions can be transferable between each other. This dissertation look into two video-related tasks in computer vision and propose a novel problem in computational astronomy.

Specifically, acquiring an in-depth understanding of videos has been a cornerstone problem in computer vision. This problem has been studied by various researchers from different perspectives, among which video prediction has attracted much attention. Video prediction aims to generate the pixels of future frames given a sequence of context frames. …

Novel Statistical Modeling Methods For Traffic Video Analysis, Hang Shi

Dissertations

Video analysis is an active and rapidly expanding research area in computer vision and artificial intelligence due to its broad applications in modern society. Many methods have been proposed to analyze the videos, but many challenging factors remain untackled. In this dissertation, four statistical modeling methods are proposed to address some challenging traffic video analysis problems under adverse illumination and weather conditions.

First, a new foreground detection method is presented to detect the foreground objects in videos. A novel Global Foreground Modeling (GFM) method, which estimates a global probability density function for the foreground and applies the Bayes decision rule …

Computer Vision Applications For Autonomous Aerial Vehicles, Burak Kakillioglu

Dissertations - ALL

Undoubtedly, unmanned aerial vehicles (UAVs) have experienced a great leap forward over the last decade. It is not surprising anymore to see a UAV being used to accomplish a certain task, which was previously carried out by humans or a former technology. The proliferation of special vision sensors, such as depth cameras, lidar sensors and thermal cameras, and major breakthroughs in computer vision and machine learning fields accelerated the advance of UAV research and technology. However, due to certain unique challenges imposed by UAVs, such as limited payload capacity, unreliable communication link with the ground stations and data safety, UAVs …

The Accuracy Of Artificial Intelligence (Ai) Chatbots In Telemedicine, Robert K. Swick

Journal of the South Carolina Academy of Science

No abstract provided.

Discriminative Region-Based Multi-Label Zero-Shot Learning, Sanath Narayan, Akshita Gupta, Salman Khan, Fahad Shahbaz Khan, Ling Shao, Mubarak Shah

Computer Vision Faculty Publications

Multi-label zero-shot learning (ZSL) is a more realistic counter-part of standard single-label ZSL since several objects can co-exist in a natural image. However, the occurrence of multiple objects complicates the reasoning and requires region-specific processing of visual features to preserve their contextual cues. We note that the best existing multi-label ZSL method takes a shared approach towards attending to region features with a common set of attention maps for all the classes. Such shared maps lead to diffused attention, which does not discriminatively focus on relevant locations when the number of classes are large. Moreover, mapping spatially-pooled visual features to …

Laser Surface Treatment And Laser Powder Bed Fusion Additive Manufacturing Study Using Custom Designed 3d Printer And The Application Of Machine Learning In Materials Science, Hao Wen

LSU Doctoral Dissertations

Selective Laser Melting (SLM) is a laser powder bed fusion (L-PBF) based additive manufacturing (AM) method, which uses a laser beam to melt the selected areas of the metal powder bed. A customized SLM 3D printer that can handle a small quantity of metal powders was built in the lab to achieve versatile research purposes. The hardware design, electrical diagrams, and software functions are introduced in Chapter 2. Several laser surface engineering and SLM experiments were conducted using this customized machine which showed the functionality of the machine and some prospective fields that this machine can be utilized. Chapter 3 …

Research On Hybrid Deployment Strategy And Model For Key Position Air-Defense Based On Multi-Weapon Platforms, Jiaqing Wan, Pengfei Wang, Junlin Tang, Zhang Dong, Xinguo Li

Journal of System Simulation

Abstract: Multi-platform cooperative air-defense is the development direction of air-defense and antimissile warfare, and the hybrid deployment strategy and model are the important research content. Based on the background of key position air-defense research, the deployment strategy, principle and form of multi-platform cooperative air-defense are studied. The operational scenario and operational process of typical air defense are established. The principles, influencing factors and deployment methods of hybrid deployment are studied. A typical hybrid deployment strategy is proposed and a mathematical model is established. It provides support for the multi platform cooperative air defense and key position defense deployment.

Research On Integrated Optimization Approach For Car-Sharing Systems, Tang Jie, Jinxin Cao

Journal of System Simulation

Abstract: Effective scheduling and routing of employees and vehicles determines the efficiency of car-sharing systems. Aiming at the scheduling of shared cars within one day, with the objective of minimizing the total system costs and personnel costs, a bi-level optimization model for multiple traveling salesman problem with time windows is established. A genetic algorithm with multi-chromosome coding and the optimized complex mutation operator are developed for the problem solution. From the comprehensive computational experiments, it can be concluded that the total numbers of vehicles and employees with the joint routing plans satisfying the order constraints can be obtained in …

Relationship Between Suspension Damping And Stability Of Vehicle Hunting Motion, Yan Yong, Zeng Jing, Kun Xu, Feiyan Zhao

Journal of System Simulation

Abstract: In order to avoid or restrain the primary hunting stability of rail vehicles, correlation between the primary hunting stability and suspension damping parameters under the damping ratio of 0 and 5% is calculated based on the analysis of suspension parameters on the vehicle modal frequency. The method to improve the stability of vehicle hunting motion by optimizing suspension parameters is obtained. The results show that when selecting different damping ratios to calculate the critical stability, the range of damping parameters varies greatly. The lateral damping parameter in a certain range or a larger vertical damping is beneficial to keep …

Modeling On Anti-Uav System-Of-Systems Combat Ooda Loop Based On Netlogo, Zhao Zhu, Wang Yi, Ruifeng Fan, Liya Li, Qi Meng

Journal of System Simulation

Abstract: Aiming at the threats from unmanned aerial vehicle (UAV) or swarm, as well as the difficult problems of systems confrontation modeling, overall process design, and operational effectiveness evaluation in anti-UAV system-of-systems, the structure and operational process both of the UAV and anti-UAV systems are analyzed respectively, and the anti-UAV Observe-Orient-Decide-Act (OODA) system-of-systems combat model which based on multi-agent modeling platform NetLogo is proposed. Utilizing the emergence of agent role in this simulation model, the influence exerted by OODA on anti-UAV is studied by the multi-agent simulation. The results show that the OODA loop is one …

The Allocation Of Jamming Resources Based On Double Q-Learning Algorithm, Xingyuan Huang, Yanyi Li

Journal of System Simulation

Abstract: In modern warfare, the multifunctional trend of radars, even multiple radars detecting targets together, enhances the anti-jamming capability of radars. However, the traditional jamming system still follows a fixed jamming strategy, and the real-time performance of decision-making facing large numbers of radars is poor. And the cognitive jamming study is urgent. The concept of reinforcement learning is explained and the difference between Q learning algorithm and double Q learning algorithm is compared. The reinforcement learning algorithm is used to establish a model based on cognitive electronic warfare to realize the allocation of radar jamming strategies. The simulation of the …

Application Of Trust Driven Adaptive Cooperative Control Algorithm, Xindong Gui, Hongjiang Ji, Lingling Fan, Shida Liu

Journal of System Simulation

Abstract: When a multi-manipulator performs cooperative task, it's end position is difficult to accurately track the target and ensure the consistency influenced by the nonlinear factors such as working environment, assembly condition and system disturbances, resulting in large errors during collaborative work. To solve multi-manipulators position coordinated control problem, a multi-agent based adaptive position coordination control algorithm is proposed by researching on the novel trust mechanism, including adaptive update mechanism of trust value (self-trust and mutual-trust) and Fisher information weighted (covariance) update mechanism. In automated collaborative assembly tasks, the precise consensus positioning of end-effector is achieved with improvements in the …