A5037185691- Distributed and Parallel Computing Systems
- Simulation and Modeling Applications
- Cloud Computing and Resource Management
- Simulation Techniques and Applications
- Reinforcement Learning in Robotics
- Industrial Vision Systems and Defect Detection
- Peer-to-Peer Network Technologies
- Human Motion and Animation
- Manufacturing Process and Optimization
- Network Security and Intrusion Detection
- Advanced Malware Detection Techniques
- Advanced Statistical Process Monitoring
- Domain Adaptation and Few-Shot Learning
- Delphi Technique in Research
- Scientific Computing and Data Management
- Surgical Simulation and Training
- Internet Traffic Analysis and Secure E-voting
- Machine Learning and Algorithms
- Machine Learning and Data Classification
- Multi-Agent Systems and Negotiation
- Smart Grid Security and Resilience
- Artificial Intelligence in Games
- Risk and Safety Analysis
- Multimodal Machine Learning Applications
- Caching and Content Delivery
University of Central Florida
2009-2025
St. Thomas University
2018-2020
DEVCOM Army Research Laboratory
2015-2016
Nuclear power plants increasingly utilize digitalized systems, including computer-based procedures (CBPs) and automation. These novel technologies require human factors’ evaluation to ensure safety. Potentially, automation contributes safety by reducing workload, but may also induce a loss of situation awareness trust miscalibration. The current study investigated workload during simulated nuclear plant (NPP) emergency operation procedure (EOP) executed using CBP supported Two levels (LOA)...
Machinery within semiconductor manufacturing facilities exhibits high degrees of integration and complexity, making it susceptible to unexpected failures. These failures pose the risk disrupting all operations, indirectly impacting profit overall production output. To avoid such problems, predictive models can be used forecast next tool state anticipate failure. Using these predictions, preventative maintenance measures take place disruptions supply chain. This paper introduces two...
This paper introduces a novel application of Generative Adversarial Networks (GANs) in the creation digital twin semiconductor manufacturing plant, as well investigation new evaluation metrics synthetic data quality. Digital twins are next technological step for long-term capacity planning, performance optimization, and job scheduling. However, real-time predictive analytics this industry demands substantial historical about equipment, operations, like Overall Equipment Effectiveness (OEE)....
Layered Learning is an iterative machine learning technique used to train agents how perform tasks. The decomposes a task into simpler components and trains the agent learn progressively more complex sub-tasks solve overall task. has been successfully instruct computer programs Boolean-logic problems, teach robots walk, RoboCup soccer playing agents. proposed work answers question of well does apply evolved development heterogeneous team Non-playable Characters (NPCs) in video game. compares...
In this paper, we describe the analysis of effect vertical computational scaling on performance a simulation based training prototype currently under development by U.S. Army Research Laboratory. The United States military is interested in facilitating Warfighter invest
The United States military is investigating large-scale, realistic virtual world simulations to facilitate warfighter training. As the simulation community strives towards meeting these training objectives, methods must be developed and validated that measure scalability performance in simulators. With such methods, will able quantifiably compare between system changes. This work contributes development validation prerequisite by evaluating effectiveness of commonly used metrics a...
Layered learning is a machine paradigm used in conjunction with direct-policy search reinforcement methods to find high performance agent behaviors for complex tasks. At its core, layered decomposition-based that shares many characteristics robot shaping, transfer learning, hierarchical decomposition, and incremental learning. Previous studies have provided evidence has the ability outperform standard monolithic of cases. The dilemma balancing stability plasticity common problem causes...
Catastrophic forgetting is a stability–plasticity imbalance that causes machine learner to lose previously gained knowledge critical for performing task. The occurs in transfer learning, negatively affecting the learner’s performance, particularly neural networks and layered learning. This work proposes complementary learning technique introduces long- short-term memory reduce negative effects of catastrophic forgetting. In particular, this dual system non-neural network approaches...
Internet of Things (IoT) devices are common in today’s computer networks. These can be computationally powerful, yet prone to cybersecurity exploitation. To remedy these growing security weaknesses, this work proposes a new artificial intelligence method that makes IoT networks safer through the use autonomous, swarm-based penetration testing. In work, introduced Particle Swarm Optimization (PSO) testing technique is compared against traditional linear and queue-based approaches find...
Virtual world simulation offers tremendous potential opportunities to improve, and optimize, individual collective echelon training for the military. The US Army Research Laboratory (ARL) Military Open Simulator Enterprise Strategy (MOSES) project’s charter is investigate simulation-based technology use in military specific domains. Of particular interest are attributes of virtual worlds such as geographically distributed large-scale trainee support. We have initiated a series experiments...
High-Performance Mass Spectrometry Liquid Chromatography (MS-HPLC) data analysis has led to insights within the fields of metabolomics and guides researchers potential genes interest for future cancer research. Current MS-HPLC classification techniques are time-intensive require human supervision. Statistical automated approaches, such as denoising feature detection, have been explored with moderate success.An approach that yet be adequately evaluated is through artificial neural networks....
The complexity and computational resource demands of large-scale virtual world simulations are increasing with advancements in CPU GPU computing power, availability broadband internet connectivity, reduction hardware costs. From these areas growth, cloud-based services gaining popularity for hosting due to their cost-effectiveness, ease management scale, data center virtually every region the world. next phase simulation will address fallacies over/under-utilization compute resources by...
Virtual world simulation for military training is an emerging domain. As such, detailed analysis required to optimize the performance simulators. Unfortunately, due a lack of extensive virtual analysis, simulator administrators often make arbitrary resource allocations support their environments and scenarios. In this paper, we provide lightweight predictive model that will be used in automated, dynamic allocation system popular three-dimensional open-sourced OpenSimulator. Prior...
The U.S. Army Research Laboratory (ARL) is investigating technologies and methods to enhance the next generation of tactical simulation-based trainers. A primary research objective increase number simultaneous Soldiers that can train collaborate in a shared, virtual environment. Current programs record cannot support Department Army's goal at company echelon (200 Soldiers) environment are limited platoon (42 concurrent trainees. ARL has identified scalability limiting factors be simulator's...
Abstract Bark extracts, derived from the Brazilian peppertree (Schinus terebinthifolius) were tested on BT549 triple negative breast cancer cells, a model system for invasive, metastatic cancers. Crude prepared with 50% ethanol-50% water solvent, investigated cytotoxicity and effects cell migration in vitro. Extract-treated cells exhibited statistically significant slower velocities than untreated control cells. bark screened at concentration of 3.15mg/mL five independent experiments,...
This work investigates the behavior of a distributed team agents on dynamic task allocation problem. Previous finds that decision making process can effectively assign tasks appropriately to members even when have only local information. We study this problem in environment which move, thus causing neighborhoods change over time. Results indicate higher level adaptation is clearly required environment. Despite increased difficulty, able achieve comparable both static and environments.