Physical Sciences and Mathematics Commons^™

Transition-Informed Reinforcement Learning For Large-Scale Stackelberg Mean-Field Games., Pengdeng Li, Runsheng Yu, Xinrun Wang, Bo An

Research Collection School Of Computing and Information Systems

Many real-world scenarios including fleet management and Ad auctions can be modeled as Stackelberg mean-field games (SMFGs) where a leader aims to incentivize a large number of homogeneous self-interested followers to maximize her utility. Existing works focus on cases with a small number of heterogeneous followers, e.g., 5-10, and suffer from scalability issue when the number of followers increases. There are three major challenges in solving large-scale SMFGs: i) classical methods based on solving differential equations fail as they require exact dynamics parameters, ii) learning by interacting with environment is data-inefficient, and iii) complex interaction between the leader and followers …

Earnhft: Efficient Hierarchical Reinforcement Learning For High Frequency Trading, Molei Qin, Shuo Sun, Wentao Zhang, Haochong Xia, Xinrun Wang, Bo An

Research Collection School Of Computing and Information Systems

High-frequency trading (HFT) is using computer algorithms to make trading decisions in short time scales (e.g., second-level), which is widely used in the Cryptocurrency (Crypto) market, (e.g., Bitcoin). Reinforcement learning (RL) in financial research has shown stellar performance on many quantitative trading tasks. However, most methods focus on low-frequency trading, e.g., day-level, which cannot be directly applied to HFT because of two challenges. First, RL for HFT involves dealing with extremely long trajectories (e.g., 2.4 million steps per month), which is hard to optimize and evaluate. Second, the dramatic price fluctuations and market trend changes of Crypto make existing algorithms …

Leveraging Llms And Generative Models For Interactive Known-Item Video Search, Zhixin Ma, Jiaxin Wu, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

While embedding techniques such as CLIP have considerably boosted search performance, user strategies in interactive video search still largely operate on a trial-and-error basis. Users are often required to manually adjust their queries and carefully inspect the search results, which greatly rely on the users’ capability and proficiency. Recent advancements in large language models (LLMs) and generative models offer promising avenues for enhancing interactivity in video retrieval and reducing the personal bias in query interpretation, particularly in the known-item search. Specifically, LLMs can expand and diversify the semantics of the queries while avoiding grammar mistakes or the language barrier. In …

Delving Into Multimodal Prompting For Fine-Grained Visual Classification, Xin Jiang, Hao Tang, Junyao Gao, Xiaoyu Du, Shengfeng He, Zechao Li

Research Collection School Of Computing and Information Systems

Fine-grained visual classification (FGVC) involves categorizing fine subdivisions within a broader category, which poses challenges due to subtle inter-class discrepancies and large intra-class variations. However, prevailing approaches primarily focus on uni-modal visual concepts. Recent advancements in pre-trained vision-language models have demonstrated remarkable performance in various high-level vision tasks, yet the applicability of such models to FGVC tasks remains uncertain. In this paper, we aim to fully exploit the capabilities of cross-modal description to tackle FGVC tasks and propose a novel multimodal prompting solution, denoted as MP-FGVC, based on the contrastive language-image pertaining (CLIP) model. Our MP-FGVC comprises a multimodal prompts …

Artificial Intelligence Model Predicts Sudden Cardiac Arrest Manifesting With Pulseless Electric Activity Versus Ventricular Fibrillation, Lauri Holmstrom, Bryan Bednarski, Harpriya Chugh, Habiba Aziz, Hoang Nhat Pham, Arayik Sargsyan, Audrey Uy-Evanado, Damini Dey, Angelo Salvucci, Jonathan Jui, Kyndaron Reinier, Piotr J Slomka, Sumeet S Chugh

Student and Faculty Publications

BACKGROUND: There is no specific treatment for sudden cardiac arrest (SCA) manifesting as pulseless electric activity (PEA) and survival rates are low; unlike ventricular fibrillation (VF), which is treatable by defibrillation. Development of novel treatments requires fundamental clinical studies, but access to the true initial rhythm has been a limiting factor.

METHODS: Using demographics and detailed clinical variables, we trained and tested an AI model (extreme gradient boosting) to differentiate PEA-SCA versus VF-SCA in a novel setting that provided the true initial rhythm. A subgroup of SCAs are witnessed by emergency medical services personnel, and because the response time is …

Predictive Algorithm For Surgery Recommendation In Thoracolumbar Burst Fractures Without Neurological Deficits, Charlotte Dandurand, Nader Fallah, Cumhur F. Öner, Richard J. Bransford, Klaus Schnake, Alex R. Vaccaro, Lorin M. Benneker, Emiliano Vialle, Gregory D. Schroeder, Shanmuganathan Rajasekaran, Mohammad El-Skarkawi, Rishi M. Kanna, Mohamed Aly, Martin Holas, Jose A. Canseco, Sander Muijs, Eugen Cezar Popescu, Jin Wee Tee, Gaston Camino-Willhuber, Andrei Fernandes Joaquim, Ory Keynan, Harvinder Singh Chhabra, Sebastian Bigdon, Ulrich Spiegel, Marcel F. Dvorak

Department of Orthopaedic Surgery Faculty Papers

STUDY DESIGN: Predictive algorithm via decision tree.

OBJECTIVES: Artificial intelligence (AI) remain an emerging field and have not previously been used to guide therapeutic decision making in thoracolumbar burst fractures. Building such models may reduce the variability in treatment recommendations. The goal of this study was to build a mathematical prediction rule based upon radiographic variables to guide treatment decisions.

METHODS: Twenty-two surgeons from the AO Knowledge Forum Trauma reviewed 183 cases from the Spine TL A3/A4 prospective study (classification, degree of certainty of posterior ligamentous complex (PLC) injury, use of M1 modifier, degree of comminution, treatment recommendation). Reviewers' regions …

Deep Learning-Based Human Action Understanding In Videos, Elahe Vahdani

Dissertations, Theses, and Capstone Projects

The understanding of human actions in videos holds immense potential for technological advancement and societal betterment. This thesis explores fundamental aspects of this field, including action recognition in trimmed clips and action localization in untrimmed videos. Trimmed videos contain only one action instance, with moments before or after the action excluded from the video. However, the majority of videos captured in unconstrained environments, often referred to as untrimmed videos, are naturally unsegmented. Untrimmed videos are typically lengthy and may encompass multiple action instances, along with the moments preceding or following each action, as well as transitions between actions. In the …

Does Chatgpt Know Calculus?, Kris H. Green

Journal of Humanistic Mathematics

Academics and educators across the world are grappling with how OpenAI’s new software, ChatGPT, will impact teaching and learning. This essay explores ChatGPT’s response to a typical calculus problem as a way of illustrating its functionality and limitations.

Dynamic Prognosis Prediction For Patients On Dapt After Drug-Eluting Stent Implantation: Model Development And Validation, Fang Li, Laila Rasmy, Yang Xiang, Jingna Feng, Ahmed Abdelhameed, Xinyue Hu, Zenan Sun, David Aguilar, Abhijeet Dhoble, Jingcheng Du, Qing Wang, Shuteng Niu, Yifang Dang, Xinyuan Zhang, Ziqian Xie, Yi Nian, Jianping He, Yujia Zhou, Jianfu Li, Mattia Prosperi, Jiang Bian, Degui Zhi, Cui Tao

School of Medicine Faculty Publications

BACKGROUND: The rapid evolution of artificial intelligence (AI) in conjunction with recent updates in dual antiplatelet therapy (DAPT) management guidelines emphasizes the necessity for innovative models to predict ischemic or bleeding events after drug-eluting stent implantation. Leveraging AI for dynamic prediction has the potential to revolutionize risk stratification and provide personalized decision support for DAPT management. METHODS AND RESULTS: We developed and validated a new AI-based pipeline using retrospective data of drug-eluting stent-treated patients, sourced from the Cerner Health Facts data set (n=98 236) and Optum’s de-identified Clinformatics Data Mart Database (n=9978). The 36 months following drug-eluting stent implantation were …

Promises And Risks Of Applying Ai Medical Imaging To Early Detection Of Cancers, And Regulation For Ai Medical Imaging, Yiyao Zhang

The Journal of Purdue Undergraduate Research

No abstract provided.

A Survey On Training Challenges In Generative Adversarial Networks For Biomedical Image Analysis, Muhammad Muneeb Saad, Ruairi O'Reilly, Mubashir Husain Rehmani

Department of Computer Science Publications

In biomedical image analysis, the applicability of deep learning methods is directly impacted by the quantity of image data available. This is due to deep learning models requiring large image datasets to provide high-level performance. Generative Adversarial Networks (GANs) have been widely utilized to address data limitations through the generation of synthetic biomedical images. GANs consist of two models. The generator, a model that learns how to produce synthetic images based on the feedback it receives. The discriminator, a model that classifies an image as synthetic or real and provides feedback to the generator. Throughout the training process, a GAN …

De Novo Drug Design Using Transformer-Based Machine Translation And Reinforcement Learning Of An Adaptive Monte Carlo Tree Search, Dony Ang, Cyril Rakovski, Hagop S. Atamian

Biology, Chemistry, and Environmental Sciences Faculty Articles and Research

The discovery of novel therapeutic compounds through de novo drug design represents a critical challenge in the field of pharmaceutical research. Traditional drug discovery approaches are often resource intensive and time consuming, leading researchers to explore innovative methods that harness the power of deep learning and reinforcement learning techniques. Here, we introduce a novel drug design approach called drugAI that leverages the Encoder–Decoder Transformer architecture in tandem with Reinforcement Learning via a Monte Carlo Tree Search (RL-MCTS) to expedite the process of drug discovery while ensuring the production of valid small molecules with drug-like characteristics and strong binding affinities towards …

Anthropomorphism And Human-Robot Interaction, Rae Yule Kim

Department of Economics Faculty Scholarship and Creative Works

Exploring how human appreciation for and interactions with robots are influenced by anthropomorphic features.

A Target-Based And A Targetless Extrinsic Calibration Methods For Thermal Camera And 3d Lidar, Farhad Dalirani

Electronic Thesis and Dissertation Repository

This thesis introduces two novel methods for the extrinsic calibration of a thermal camera and a 3D LiDAR sensor, which are crucial for seamless data integration. The first method employs a distinctive calibration target, leveraging lines and plane equations correspondence in both modalities for a single pose, and incorporating more poses by matching the target's edges. It achieves reliable results, even with just one pose yielding 10.82% translation and 0.51-degree rotation errors. This outperforms alternative methods, which require eight pairs for similar results. The second method eliminates the need for a dedicated target. Instead, by collecting data during the sensor …

Scene Graph Generation: A Comprehensive Survey, Hongsheng Li, Guangming Zhu, Liang Zhang, Youliang Jiang, Yixuan Dang, Haoran Hou, Peiyi Shen, Xia Zhao, Syed A. A. Shah, Mohammed Bennamoun

Research outputs 2022 to 2026

Deep learning techniques have led to remarkable breakthroughs in the field of object detection and have spawned a lot of scene-understanding tasks in recent years. Scene graph has been the focus of research because of its powerful semantic representation and applications to scene understanding. Scene Graph Generation (SGG) refers to the task of automatically mapping an image or a video into a semantic structural scene graph, which requires the correct labeling of detected objects and their relationships. In this paper, a comprehensive survey of recent achievements is provided. This survey attempts to connect and systematize the existing visual relationship detection …

Obstacle Avoidance Motion In Mobile Robotics, Yunchao Tang, Shaojun Qi, Lixue Zhu, Xianrong Zhuo, Yunqi Zhang, Fan Meng

Journal of System Simulation

Abstract: The advancement of artificial intelligence technology has significantly enhanced the utilization of mobile robots in various fields such as industry, aerospace, and agriculture. The autonomous obstacle avoidance capability of these robots is crucial to the safety and efficiency of their operations in diverse settings. Path planning, a key technology in obstacle avoidance, plays an essential role in the overall performance of these systems. This paper presents a comprehensive review of path planning technology for mobile robots, categorizing the algorithms into global planning and local obstacle avoidance according to their operational requirements. Specific focus is given to the global planning …

Multi-Uav Collaborative Trajectory Planning Algorithm For Urban Ultra-Low-Altitude Air Transportation Scenario, Jie Cheng, Yuan Zheng, Chenglong Li, Bo Jiang

Journal of System Simulation

Abstract: The rapid development of the drone industry has promoted the opening of low-altitude, forming a wave of ultra-low-altitude air transportation in cities sweeping over the world. However, the existing trajectory planning algorithms do not consider the division method and operating rules of the ultra-low-altitude airspace. They are not suitable for the collaborative trajectory planning of multiple UAVs in the urban ultra-low-altitude air transportation scenario, which may restrict the development of the ultra-low-altitude air transportation industry. This paper explores a multi-UAV collaborative trajectory planning method for urban ultra-low-altitude air transportation scenario based on the airspace flight altitude layer architecture. Specifically, …

Action Recognition Model Of Directed Attention Based On Cosine Similarity, Chen Li, Ming He, Chen Dong, Wei Li

Journal of System Simulation

Abstract: Aiming at the lack of directionality of traditional dot product attention, this paper proposes a directed attention model (DAM) based on cosine similarity. To effectively represent the direction relationship between the spatial and temporal features of video frames, the paper defines the relationship function in the attention mechanism using the cosine similarity theory, which can remove the absolute value of the relationship between features. To reduce the computational burden of the attention mechanism, the operation is decomposed from two dimensions of time and space. The computational complexity is further optimized by combining linear attention operation. The experiment is divided …

Overall Scheme Design And Integration Testing Of Hardware-In-The-Loop Simulation Of Guidance And Control System, Xiaofei Chang, Jiayue Jiao, Kang Chen, Wenxing Fu, Jie Yan

Journal of System Simulation

Abstract: Hardware-in-the-loop simulation system is a complex distributed simulation system, and its design and integration directly affect the system performance and construction goals. Based on years of experience, this paper first summarizes the design of the overall scheme and analyzes the performance requirements of real-time, compatibility, scalability, and security. Then, the paper describes the overall scheme of a typical hardware-in-the-loop simulation system, including the functional hierarchy, operation mechanism, and structural composition. Finally, it summarizes the contents and steps of the system integration testing, acceptance testing, and credibility evaluation method.

Emergency Material Scheduling Based On Discrete Shuffled Frog Leaping Algorithm, Xiaoning Shen, Zhongpei Ge, Chengbin Yao, Liyan Song, Yufang Wang

Journal of System Simulation

Abstract: A mathematical model of emergency material scheduling after earthquakes is built. The model evaluates the emergency degree of each disaster area based on the disaster situation and designs a method to split the demand of the disaster area, improving the efficiency of vehicle utilization. To solve the model, this paper proposes a discrete shuffled frog leaping algorithm with multi-resource learning. The multiple information sources introduced by the proposed algorithm can expand the search direction and reduce the assimilation speed of the population in the algorithm. Second, the worst individual in each subgroup can learn the effective information in the …

Multi-View Depth Estimation Based On Adaptive Space Feature Enhancement, Dong Wei, Huan Liu, Xiaohan Zhang, Changkai Li, Tianyi Sun, Ziyou Zhang

Journal of System Simulation

Abstract: A multi-view depth estimation algorithm based on adaptive space feature enhancement (ASFE) is presented to improve the multi-view depth estimation accuracy. A multi-scale feature extraction module composed of an improved feature pyramid network (FPN) and ASFE is designed. This module obtains multi-scale feature maps with global context-aware information and coordinate information. The residual learning network is used to optimize the depth map to prevent the problem of blurred reconstructed edges in multiple convolution operations. The proposed algorithm constructs a focal loss function through the idea of classification to enhance the prediction ability of the network model. The experimental results …

Visual Monitoring System Of Digital Twin Workshop For Process Manufacturing, Yanchao Yin, Jiasheng Feng, Bin Yi, Wang Li, Qingwen Yin

Journal of System Simulation

Abstract: To solve the problem of poor real-time and accuracy of process manufacturing monitoring, this paper proposes a visual monitoring system architecture of digital twin workshop based on production factor data, production process, and multi-process equipment entity object amid the rapid development of digital twin technology and increasingly close integration with the manufacturing industry. Twin modeling is conducted on multi-process coupling production line entity objects from multiple dimensions such as element, state, and operation logic. Based on key technologies of data collection and transmission of OPC UA (OLE for process control unified archiecture) unified architecture and real-time drive of the …

Reentrant Hybrid Flow Shop Scheduling Problem Based On Moma, Hongbin Qin, Chenxiao Li, Hongtao Tang, Feng Zhang

Journal of System Simulation

Abstract: For the characteristics of multi-variety, large-scale and mixed-flow production of reentrant manufacturing systems, the reentrant hybrid flow shop scheduling problem with batch processors (BPRHFSP) is constructed, and an improved multi-objective mayfly algorithm (MOMA) is proposed for BPRHFSP. Firstly, decoding rules for single-piece processing stage and batch-processing stage are proposed. Then, a reverse learning initialization strategy based on logistic chaotic mapping is designed to improve the quality of the initial solution of the algorithm, also an improved mayfly mating strategy is designed to improve the local search ability of MOMA. Finally, a VND-based mayfly movement strategy is designed based on …

A Simulation Method Based On Multi-Source Sensors For Aircraft Type Identification, Shaozhu Gu, Yuxin Ying, Huajie Zhang, Yiqi Tong

Journal of System Simulation

Abstract: Existing simulation methods for aircraft type identification mainly focus on a single sensor and a single target. They do not consider the joint acquisition of aircraft parameters by various sensor devices such as optoelectronics, radar, and electronic detection in real scenarios, leading to the simple simulation scenarios. This paper proposes a simulation platform based on multi-source sensors. Specifically, the platform includes an infrared image simulator that uses a cycleGAN network to generate infrared images of the aircraft, a flight simulator that adopts the three-degree-of-freedom flight control method to generate the movement trajectory of the aircraft, a radar simulator, that …

Combat Effectiveness Evaluation Method Of Homogeneous Cluster Equipment System Based On Rlomag+Eas, Guohui Zhang, Ang Gao, Ya'nan Zhang

Journal of System Simulation

Abstract: The equipment system is the reflection of the combat system from the perspective of equipment. The research on the combat effectiveness evaluation of the equipment system is of great practical significance for the optimization, construction, and development of the equipment system. Cluster equipment combat system confrontation is characterized by large-scale, highly dynamic and strong confrontation, and it is difficult to directly evaluate combat effectiveness with traditional methods. Aiming at the single task homogeneous cluster equipment system (such as UAV reconnaissance swarm and ground unmanned platform fire assault cluster), this paper regards the confrontation process of equipment system as the …

Research On Campus Epidemic Evolution Based On Multi-Scale Modeling And Simulation In Microscopic & Microscopic View, Mingwei Hu, Wenjie Yang

Journal of System Simulation

Abstract: High density of population leads to high possibility of cross-infection. It is necessary to focus on campus epidemic prevention and control. Basing on existing studies in macroscopic or microscopic view, this paper proposed a multi-scale means to analyze a short-term evolution of Corona virus disease 2019 (COVID-19) on campus and estimated the efficiency of prevention strategies. Macroscopic model was based on the susceptible-exposed-infections-recovered(SEIR) model, which exported the time curve of the number of asymptomatic patients and symptomatic patients. Microscopic model combined discrete event simulation modeling and agent-based modeling to simulate the behavior of campus students and the state evolution …

Recursive Subspace-Based Model Refinement Method For Digital Twin Of Thermal Power Unit, Yanbo Zhao, Yuanli Cai, Huaizhong Hu

Journal of System Simulation

Abstract: Due to factors such as simplified assumptions or equipment characteristic deviation, modeling errors are inevitable in the mechanism modeling of thermal power units. To deal with the problem, this paper proposes a novel model refinement method based on recursive subspace for the digital twin of thermal power units. Firstly, the digital twin models are built based on mechanism analysis and combined with small sample data of typical conditions, ensuring interpretability and generalization performance. Secondly, based on the recursive subspace identification method, the refinement model is built and updated online in real time to compensate for the modeling error, improving …

Modeling And Analysing Of Complex Combat Systems Based On Symbiosis Theory, Xiangrui Tian, Jie Ying, Rui Yao, Xiaodong Wan

Journal of System Simulation

Abstract: As the combat systems develop towards clustering, coordination, unmanned, and intelligent direction, traditional combat system modeling methods are unable to reflect the complexity and intelligence of the combat systems. By drawing on symbiosis theory, this paper models and analyzes complex combat systems, and decomposes the complex combat system into various subsystems according to combat missions. The combat units, interaction modes, and combat environments in the subsystems are analyzed. The paper builds mathematical models to capture the collaborative interaction relationships between combat units, finally constructing a symbiotic model of complex combat systems. By the symbiosis principles and methods, quantitative analysis …

Driving Method Of Virtual Multi-Person Disassembly And Assembly Task For Aeroengine, Qiuwei Zeng, Zhaoyong Hu, Zhile Wang, Ruilin Zhang, Gang Zou

Journal of System Simulation

Abstract: To meet the needs of virtual multi-person collaborative disassembly and assembly system for different disassembly and assembly tasks, this paper proposes a data-driven method with configurable task sequence. Taking an aeroengine prototype as the research object, the paper studies the task elements of multi-person collaborative disassembly and assembly. It parameterizes and expresses the task based on JSON (JavaScript object notation) and drives the task sequence by JSON parametrical files, defining the interactive operation of each task step. The practice shows that this method is applied to the multi-person collaborative disassembly and assembly system, which makes the system configurable and …

Heterogeneous Multi-Ant Colony Algorithm Combining Competitive Interaction Strategy And Eliminatingreconstructing Mechanism, Chen Feng, Xiaoming You, Sheng Liu

Journal of System Simulation

Abstract: The traditional ant colony algorithm has many problems in convergence and diversity when solving the traveling salesman problem (TSP). Therefore, this paper proposes a heterogeneous multi-ant colony algorithm that combines the competitive interaction strategy and the eliminating-reconstructing mechanism (CEACO) to overcome these shortcomings. Firstly, the algorithm uses a competitive interaction strategy, which adjusts the interaction period adaptively according to the Hamming distance of different groups in different periods. Competition coefficients are adopted to differentiate matching interaction objects for interaction. The matched objects interact with each other through the optimal solution and pheromone matrix. This mechanism achieves a balance between …