Engineering Commons^™

The Troubling Emergence Of Hallucination In Large Language Models--An Extensive Definition, Quantification, And Prescriptive Remediations, Vipula Rawte, Swagata Chakraborty, Agnibh Pathak, Anubhav Sarkar, S.M Towhidul Islam Tonmoy, Aman Chadha, Amit Sheth, Amitava Das

Publications

The recent advancements in Large Language Models (LLMs) have garnered widespread acclaim for their remarkable emerging capabilities. However, the issue of hallucination has parallelly emerged as a by-product, posing significant concerns. While some recent endeavors have been made to identify and mitigate different types of hallucination, there has been a limited emphasis on the nuanced categorization of hallucination and associated mitigation methods. To address this gap, we offer a finegrained discourse on profiling hallucination based on its degree, orientation, and category, along with offering strategies for alleviation. As such, we define two overarching orientations of hallucination: (i) factual mirage (FM) …

Light Auditor: Power Measurement Can Tell Private Data Leakage Through Iot Covert Channels, Woosub Jung, Kailai Cui, Kenneth Koltermann, Junjie Wang, Chunsheng Xin, Gang Zhou

Electrical & Computer Engineering Faculty Publications

Despite many conveniences of using IoT devices, they have suffered from various attacks due to their weak security. Besides well-known botnet attacks, IoT devices are vulnerable to recent covert-channel attacks. However, no study to date has considered these IoT covert-channel attacks. Among these attacks, researchers have demonstrated exfiltrating users' private data by exploiting the smart bulb's capability of infrared emission.

In this paper, we propose a power-auditing-based system that defends the data exfiltration attack on the smart bulb as a case study. We first implement this infrared-based attack in a lab environment. With a newly-collected power consumption dataset, we pre-process …

Mwirgan: Unsupervised Visible-To Mwir Image Translation With Generative Adversarial Network, Mohammad Shahab Uddin, Chiman Kwan, Jiang Li

Electrical & Computer Engineering Faculty Publications

Unsupervised image-to-image translation techniques have been used in many applications, including visible-to-Long-Wave Infrared (visible-to-LWIR) image translation, but very few papers have explored visible-to-Mid-Wave Infrared (visible-to-MWIR) image translation. In this paper, we investigated unsupervised visible-to-MWIR image translation using generative adversarial networks (GANs). We proposed a new model named MWIRGAN for visible-to-MWIR image translation in a fully unsupervised manner. We utilized a perceptual loss to leverage shape identification and location changes of the objects in the translation. The experimental results showed that MWIRGAN was capable of visible-to-MWIR image translation while preserving the object’s shape with proper enhancement in the translated images and …

Ultrasensitive Tapered Optical Fiber Refractive Index, Erem Ujah, Meimei Lai, Gymama Slaughter

Electrical & Computer Engineering Faculty Publications

Refractive index (RI) sensors are of great interest for label-free optical biosensing. A tapered optical fiber (TOF) RI sensor with micron-sized waist diameters can dramatically enhance sensor sensitivity by reducing the mode volume over a long distance. Here, a simple and fast method is used to fabricate highly sensitive refractive index sensors based on localized surface plasmon resonance (LSPR). Two TOFs (l = 5 mm) with waist diameters of 5 µm and 12 µm demonstrated sensitivity enhancement at λ = 1559 nm for glucose sensing (5-45 wt%) at room temperature. The optical power transmission decreased with increasing glucose concentration due …

The Effect Of The Width Of The Incident Pulse To The Dielectric Transition Layer In The Scattering Of An Electromagnetic Pulse — A Qubit Lattice Algorithm Simulation, George Vahala, Linda Vahala, Abhay K. Ram, Min Soe

Electrical & Computer Engineering Faculty Publications

The effect of the thickness of the dielectric boundary layer that connects a material of refractive index n₁ to another of index n₂is considered for the propagation of an electromagnetic pulse. A qubit lattice algorithm (QLA), which consists of a specially chosen non-commuting sequence of collision and streaming operators acting on a basis set of qubits, is theoretically determined that recovers the Maxwell equations to second-order in a small parameter ϵ. For very thin boundary layer the scattering properties of the pulse mimics that found from the Fresnel jump conditions for a plane wave - except that …

Toward Real-Time, Robust Wearable Sensor Fall Detection Using Deep Learning Methods: A Feasibility Study, Haben Yhdego, Christopher Paolini, Michel Audette

Electrical & Computer Engineering Faculty Publications

Real-time fall detection using a wearable sensor remains a challenging problem due to high gait variability. Furthermore, finding the type of sensor to use and the optimal location of the sensors are also essential factors for real-time fall-detection systems. This work presents real-time fall-detection methods using deep learning models. Early detection of falls, followed by pneumatic protection, is one of the most effective means of ensuring the safety of the elderly. First, we developed and compared different data-segmentation techniques for sliding windows. Next, we implemented various techniques to balance the datasets because collecting fall datasets in the real-time setting has …

Class Activation Mapping And Uncertainty Estimation In Multi-Organ Segmentation, Md. Shibly Sadique, Walia Farzana, Ahmed Temtam, Khan Iftekharuddin, Khan Iftekharuddin (Ed.), Weijie Chen (Ed.)

Electrical & Computer Engineering Faculty Publications

Deep learning (DL)-based medical imaging and image segmentation algorithms achieve impressive performance on many benchmarks. Yet the efficacy of deep learning methods for future clinical applications may become questionable due to the lack of ability to reason with uncertainty and interpret probable areas of failures in prediction decisions. Therefore, it is desired that such a deep learning model for segmentation classification is able to reliably predict its confidence measure and map back to the original imaging cases to interpret the prediction decisions. In this work, uncertainty estimation for multiorgan segmentation task is evaluated to interpret the predictive modeling in DL …

Transfer Learning Using Infrared And Optical Full Motion Video Data For Gender Classification, Alexander M. Glandon, Joe Zalameda, Khan M. Iftekharuddin, Gabor F. Fulop (Ed.), David Z. Ting (Ed.), Lucy L. Zheng (Ed.)

Electrical & Computer Engineering Faculty Publications

This work is a review and extension of our ongoing research in human recognition analysis using multimodality motion sensor data. We review our work on hand crafted feature engineering for motion capture skeleton (MoCap) data, from the Air Force Research Lab for human gender followed by depth scan based skeleton extraction using LIDAR data from the Army Night Vision Lab for person identification. We then build on these works to demonstrate a transfer learning sensor fusion approach for using the larger MoCap and smaller LIDAR data for gender classification.

Long-Range Aceo Phenomena In Microfluidic Channel, Diganta Dutta, Keifer Smith, Xavier Palmer

Electrical & Computer Engineering Faculty Publications

Microfluidic devices are increasingly utilized in numerous industries, including that of medicine, for their abilities to pump and mix fluid at a microscale. Within these devices, microchannels paired with microelectrodes enable the mixing and transportation of ionized fluid. The ionization process charges the microchannel and manipulates the fluid with an electric field. Although complex in operation at the microscale, microchannels within microfluidic devices are easy to produce and economical. This paper uses simulations to convey helpful insights into the analysis of electrokinetic microfluidic device phenomena. The simulations in this paper use the Navier–Stokes and Poisson Nernst–Planck equations solved using COMSOL …

Deep-Learning-Based Classification Of Digitally Modulated Signals Using Capsule Networks And Cyclic Cumulants, John A. Snoap, Dimitrie C. Popescu, James A. Latshaw, Chad M. Spooner

Electrical & Computer Engineering Faculty Publications

This paper presents a novel deep-learning (DL)-based approach for classifying digitally modulated signals, which involves the use of capsule networks (CAPs) together with the cyclic cumulant (CC) features of the signals. These were blindly estimated using cyclostationary signal processing (CSP) and were then input into the CAP for training and classification. The classification performance and the generalization abilities of the proposed approach were tested using two distinct datasets that contained the same types of digitally modulated signals, but had distinct generation parameters. The results showed that the classification of digitally modulated signals using CAPs and CCs proposed in the paper …

Continuity Of Formal Power Series Products In Nonlinear Control Theory, W. Steven Gray, Mathias Palmstrøm, Alexander Schmeding

Electrical & Computer Engineering Faculty Publications

Formal power series products appear in nonlinear control theory when systems modeled by Chen–Fliess series are interconnected to form new systems. In fields like adaptive control and learning systems, the coefficients of these formal power series are estimated sequentially with real-time data. The main goal is to prove the continuity and analyticity of such products with respect to several natural (locally convex) topologies on spaces of locally convergent formal power series in order to establish foundational properties behind these technologies. In addition, it is shown that a transformation group central to describing the output feedback connection is in fact an …

A Survey Of Using Machine Learning In Iot Security And The Challenges Faced By Researchers, Khawlah M. Harahsheh, Chung-Hao Chen

Electrical & Computer Engineering Faculty Publications

The Internet of Things (IoT) has become more popular in the last 15 years as it has significantly improved and gained control in multiple fields. We are nowadays surrounded by billions of IoT devices that directly integrate with our lives, some of them are at the center of our homes, and others control sensitive data such as military fields, healthcare, and datacenters, among others. This popularity makes factories and companies compete to produce and develop many types of those devices without caring about how secure they are. On the other hand, IoT is considered a good insecure environment for cyber …

Ultra-Low Intensity Post-Pulse Affects Cellular Responses Caused By Nanosecond Pulsed Electric Fields, Kamal Asadipour, Carol Zhou, Vincent Yi, Stephen J. Beebe, Shu Xiao

Electrical & Computer Engineering Faculty Publications

High-intensity nanosecond pulse electric fields (nsPEF) can preferentially induce various effects, most notably regulated cell death and tumor elimination. These effects have almost exclusively been shown to be associated with nsPEF waveforms defined by pulse duration, rise time, amplitude (electric field), and pulse number. Other factors, such as low-intensity post-pulse waveform, have been completely overlooked. In this study, we show that post-pulse waveforms can alter the cell responses produced by the primary pulse waveform and can even elicit unique cellular responses, despite the primary pulse waveform being nearly identical. We employed two commonly used pulse generator designs, namely the Blumlein …

A Review Of Iot Security And Privacy Using Decentralized Blockchain Techniques, Vinay Gugueoth, Sunitha Safavat, Sachin Shetty, Danda Rawat

Electrical & Computer Engineering Faculty Publications

IoT security is one of the prominent issues that has gained significant attention among the researchers in recent times. The recent advancements in IoT introduces various critical security issues and increases the risk of privacy leakage of IoT data. Implementation of Blockchain can be a potential solution for the security issues in IoT. This review deeply investigates the security threats and issues in IoT which deteriorates the effectiveness of IoT systems. This paper presents a perceptible description of the security threats, Blockchain based solutions, security characteristics and challenges introduced during the integration of Blockchain with IoT. An analysis of different …

Quantum Efficiency And Lifetime Study For Negative Electron Affinity Gaas Nanopillar Array Photocathode, Md Aziz Ar Rahman, Md Abdullah Mamun, Shukui Zhang, Hani E. Elsayed-Ali

Electrical & Computer Engineering Faculty Publications

Recent studies showed significant improvement in quantum efficiency (QE) by negative electron affinity (NEA) GaAs nanopillar array (NPA) photocathodes over their flat surface peers, particularly at 500 ─ 800 nm waveband. However, the underlying physics is yet to be well understood for further improvement in its performance. In this report, NEA GaAs NPA photocathodes with different dimensions were studied. The diameter of the nanopillars varied from 200 ─ 360 nm, the height varied from 230 ─ 1000 nm and the periodicity varied from 470 ─ 630 nm. The QE and photocathode lifetime were measured. Mie-resonance enhancement was observed at tunable …

Virtual Surgical Planning In Craniomaxillofacial Surgery: A Structured Review, Kaye Verlarde, Rentor Cafino, Armando Isla Jr., Karen Mae Ty, Xavier-Lewis Palmer, Lucas Potter, Larry Nadorra, Luchin Valrian Pueblos, Lemuel Clark Velasco

Electrical & Computer Engineering Faculty Publications

Craniomaxillofacial (CMF) surgery is a challenging and very demanding field that involves the treatment of congenital and acquired conditions of the face and head. Due to the complexity of the head and facial region, various tools and techniques were developed and utilized to aid surgical procedures and optimize results. Virtual Surgical Planning (VSP) has revolutionized the way craniomaxillofacial surgeries are planned and executed. It uses 3D imaging computer software to visualize and simulate a surgical procedure. Numerous studies were published on the usage of VSP in craniomaxillofacial surgery. However, the researchers found inconsistency in the previous literature which prompted the …

Technology Adoption Of Computer-Aided Instruction In Healthcare: A Structured Review, Queenie Kate Cabanilla, Frevy Teofilo-Orencia, Rentor Cafino, Armando T. Isla Jr., Jehan Grace Maglaya, Xavier-Lewis Palmer, Lucas Potter, Dave E. Marcial, Lemuel Clark Velasco

Electrical & Computer Engineering Faculty Publications

Computer-Aided Instruction (CAI) is one of the interactive teaching methods that electronically presents instructional resources and enhances learner performance. In health settings, using CAI is one of the important ways to improve learners' knowledge and usefulness in their healthcare specialization yet there is still a lack of research that offers a comprehensive synthesis of investigating into the adoption of CAI in healthcare. This research aims to provide a comprehensive review of related literatures on the enablers and barriers for technology adoption of CAI in healthcare. 31 journals were analyzed and revealed that several studies were utilizing the Unified Theory of …

High Figure Of Merit Spin Polarized Electron Sources Grown Via Mocvd, Benjamin Belfore, Adam Masters, Deewakar Poudel, Greg Blume, Stephen Polly, Erdong Wang, Seth M. Hubbard, Marcy Stutzman, Joseph M. Grames, Matt Poelker, Matt Grau, Sylvain Marsillac

Electrical & Computer Engineering Faculty Publications

Spin polarized photocathodes are key to the future operation of electron accelerators such as the ones at Thomas Jefferson National Accelerator Facility and Brookhaven National Laboratory. Currently, these photocathodes come in short supply due to limited production by molecular beam epitaxy. By developing a process to implement similar structures using metal organic chemical vapor deposition, the availability of these devices can be increased. In this paper, we detail the implementation of recent photocathode advancements via metal organic chemical vapor deposition process and show an improvement in both polarization and quantum efficiency of our devices compared to those fabricated via molecular …

Social-Engineering, Bio-Economies, And Nation-State Ontological Security: A Commentary, Brandon Griffin, Keitavius Alexander, Xavier-Lewis Palmer, Lucas Potter

Electrical & Computer Engineering Faculty Publications

Biocybersecurity is an evolving discipline that aims to identify the gaps and risks associated with the convergence of Biology (the science of life and living organisms) and cybersecurity (the science, study, and theory of cyberspace and cybernetics) to protect the bioeconomy. The biological industries’ increased reliance on digitization, automation, and computing power has resulted in benefits for the scientific community, it has simultaneously multiplied the risk factors associated with industrial espionage and the protection of data both commercial and proprietary. The sensitive and potentially destructive power of this data and its access inherently poses a risk to the national and …

Implications Of Cyberbiosecurity In Advanced Agriculture, Simone Stephen, Keitavius Alexander, Lucas Potter, Xavier-Lewis Palmer

Electrical & Computer Engineering Faculty Publications

The world is currently undergoing a rapid digital transformation sometimes referred to as the fourth industrial revolution. During this transformation, it is increasingly clear that many scientific fields are not prepared for this change. One specific area is agriculture. As the sector which creates global food supply, this critical infrastructure requires detailed assessment and research via newly developed technologies (Millett et al, 2019; Peccoud et al, 2018) . Despite its fundamental significance to modern civilization, many aspects of industrial agriculture have not yet adapted to the digital world. This is evident in the many vulnerabilities currently present within agricultural systems, …

A Reflection Of Typology And Verification Flaws In Consideration Of Biocybersecurity/Cyberbiosecurity: Just Another Gap In The Wall, Luke Potter, Kim Mossberg, Xavier Palmer

Electrical & Computer Engineering Faculty Publications

Verification is central to any process in a functional and enduring cyber-secure organization. This verification is how the validity or accuracy of a state of being is assessed (Schlick, 1936; Balci, 1998). Conversely, breakdown in verification procedures is core to the interruption of normal operations for an organization. A key problem for organizations that utilize biology as an interlock within their systems is that personnel lack sufficient ability to verify all practically relevant biological information for procedures such as a nurse logging a blood draw, or a molecular biology technician preparing agar to culture microbes for study. This has several …

Finite Element Analysis Of Compaction Load To Investigate The Stress-Deformation Behavior Of Soil Geosynthetic Composite Mass: A Case Study, Truc Phan, Meen-Wah Gui, Thang Pham

Civil Engineering Faculty Publications

When building Soil Geosynthetic Composite (SGC) walls, fill compaction is normally carried out by operating a compactor in a general direction parallel to the wall face. In other words, a moving point or area load is often used to apply a compaction load on a newly installed soil lift. Pham (2009) and Wu and Pham (2010) demonstrated that the compaction-induced stress (CIS) caused by multiple passes of a compactor moving toward or away from a section can be calculated by taking into account the compaction load applied directly above the section under consideration using a simplified stress path proposed by …

Economic And Input-Use Efficiency Of Nitrogen Management Techniques In Non-Irrigated Maize Production, Katie J. Bathke

Honors Program: Senior Projects (Public)

The efficiency of nitrogen (N) management has become a main concern in agricultural cropping systems for understanding the optimal N rate to help producers improve economically and reduce the exhaustion of natural resources environmentally. Nitrogen rates vary both temporally and spatially by the interactions of the soil environment and rainfall through a growing season. Thus, a site-specific approach can further optimize this variability with the understanding of the role soil variability plays in the N rate applied to a non-irrigated field. The objective of this study was to evaluate the various nitrogen treatments for their nutrient use efficiency and economic …

Mechanical And Morphological Characterization Of Energy Materials By Nanoindentation And Scanning Probe Microscopy, Jacob Hempel

Theses and Dissertations--Physics and Astronomy

The demand for robust materials in energy-based applications such as solid electrolytes for batteries, piezoelectric composites for mechanical energy harvesting, or new polymer materials for polymer binders in composite electrodes is on the rise. Mechanical degradation is the most common way for devices to fail in operation, thus a thorough understanding of a material’s mechanical properties is essential for application purposes.

In this dissertation, the mechanical characterization of novel boron cluster solid electrolytes is presented. Therein, we determine the elastic modulus and hardness of these compounds using instrumented indentation. A comparison is made against other solid electrolytes. It is found …

การคัดกรองความผิดปกติของรูปภาพบุคคลที่สร้างจากปัญญาประดิษฐ์, ธนพล พจนะสวนีย์

Chulalongkorn University Theses and Dissertations (Chula ETD)

ในยุคปัจจุบันความสามารถของปัญญาประดิษฐ์ถูกสร้างขึ้นและพัฒนาเพื่อสามารถสร้างงานสร้างสรรค์ต่างๆ ได้แก่ รูปภาพ วิดีโอ เสียง งานเขียน และบทความ งานวิจัยนี้มุ่งเน้นไปที่ความสามารถของปัญญาประดิษฐ์ในการผลิตงานภาพบุคคลตามคําสั่งหรือข้อความที่หลากหลาย อย่างไรก็ตาม การใช้ปัญญาประดิษฐ์ประเภทนี้ยังมีความผิดพลาดเกิดขึ้น ความผิดพลาดที่พบบ่อยที่สุดในการสร้างภาพบุคคลคือความผิดปกติจากส่วนของร่างกายหรืออวัยวะของมนุษย์เช่นอวัยวะที่ผิดรูปหรือองค์ประกอบอวัยวะขาดหายไป จากปัญหานี้ทำให้ผู้วิจัยเกิดแนวคิดที่จะออกแบบแบบจำลองสําหรับการคัดกรองความผิดปกติของภาพบุคคลที่สร้างขึ้นโดยปัญญาประดิษฐ์ งานวิจัยนี้มีวัตถุประสงค์ในการสร้างชุดข้อมูลภาพบุคคลประมาณ 100,000 ภาพที่สร้างขึ้นโดยปัญญาประดิษฐ์และพัฒนาแบบจำลองต้นแบบสําหรับคัดกรองความผิดปกติของภาพบุคคลที่สร้างขึ้นโดยปัญญาประดิษฐ์เพื่อทำให้ภาพปัญญาประดิษฐ์จะมีคุณภาพที่ดีขึ้นและมีประโยชน์มากขึ้น แบบจำลองการเรียนรู้เชิงลึกนำมาใช้เพื่อแยกประเภทภาพปัญญาประดิษฐ์ โดยแบ่งออกเป็น2 ประเภท: ภาพปกติ (Normal)และ ภาพบกพร่อง (Defect) ชุดข้อมูลที่ใช้พัฒนาแบบจำลองนี้ประกอบด้วยข้อมูลประมาณ 100,000 ภาพเพื่อให้แบบจำลองมีความแม่นยําและประสิทธิภาพที่ดี งานวิจัยนี้ใช้แบบจำลองการเรียนรู้เชิงลึก 4 ประเภทในการทดสอบเพื่อหาแบบจำลองที่เหมาะสมที่สุดสําหรับการคัดกรองความผิดปกติในภาพปัญญาประดิษฐ์ แบบจำลองนี้ใช้เป็นแบบจำลองต้นแบบในการคัดกรองคุณภาพของภาพบุคคลที่สร้างขึ้นโดยปัญญาประดิษฐ์และชุดข้อมูลทั้งหมดในงานวิจัยนี้จะถูกเผยแพร่โดยสาธารณะ

การทำนายอาชญากรรมผ่านการวิเคราะห์ร่วมกันของข้อมูลสถานีตำรวจใกล้เคียง, ศิรัสวยา หงษ์ยนต์

Chulalongkorn University Theses and Dissertations (Chula ETD)

การทำนายอาชญากรรมผ่านการวิเคราะห์ร่วมกันของข้อมูลสถานีตำรวจใกล้เคียงจะช่วยทำให้ข้อมูลมีความแม่นยำมากขึ้น ซึ่งจะเป็นประโยชน์กับสถานีตำรวจแต่ละแห่งที่จะนำผลลัพธ์ที่ได้ไปประกอบการปรับกลยุทธ์การป้องกันอาชญากรรมรวมถึงการจัดสรรกำลังพลให้กับหน่วยงานของตนเองได้มีประสิทธิภาพมากขึ้น การเรียนรู้ด้วยเครื่องกลายเป็นเครื่องมือสำคัญในการทำนายอาชญากรรมซึ่งช่วยให้มีความแม่นยำมากขึ้น ในการศึกษานี้มุ่งเน้นไปที่การทำนายจำนวนอาชญากรรมที่จะเกิดขึ้นในสถานีตำรวจปทุมวัน ด้วยการใช้ข้อมูลประวัติอาชญากรรมจากสถานีตำรวจทั่วประเทศตั้งแต่มกราคมปี 2554 ถึงกุมภาพันธ์ปี 2565 โดยใช้แบบจำลองการ K-Means เพื่อจัดกลุ่มสถานีตำรวจที่มีรูปแบบอาชญากรรมคล้ายคลึงกับสถานีตำรวจปทุมวัน ผลลัพธ์ที่ได้จากการจัดกลุ่มพบว่าสถานีตำรวจวังทองหลาง, นางเลิ้ง, ดุสิต, บางซื่อ, ทุ่งมหาเมฆ, สำเหร่, สุทธิสาร, ปากคลองสาน, บางกอกใหญ่, บางกอกน้อย, มักกะสัน, บางยี่เรือและตลาดพลูถูกจัดให้อยู่กลุ่มเดียวกับสถานีตำรวจปทุมวัน จากนั้นจึงใช้แบบจำลอง LSTM ในการรทำนายอาชญากรรมของสถานีตำรวจปทุมวันผ่านการวิเคราะห์ร่วมกับข้อมูลสถานีตำรวจใกล้เคียง ชุดข้อมูลการฝึกใช้ข้อมูลสถานีตำรวจปทุมวันจับคู่กับสถานีตำรวจภายในกลุ่มเดียวกัน ซึ่งผลการวิจัยพบว่าการวิเคราะห์ร่วมกันระหว่างสถานีตำรวจปทุมวันและสถานีตำรวจวังทองหลาง, นางเลิ้ง, ดุสิต, บางซื่อ, ปากคลองสาน, บางกอกน้อย, มักกะสันและตลาดพลูส่งผลให้ RMSE และ MAE ลดลง เมื่อเทียบกับการใช้ข้อมูลสถานีตำรวจจปทุมวันเพียงอย่างเดียว แสดงให้เห็นว่าการวิเคราะห์ร่วมกันของข้อมูลสถานีตำรวจใกล้เคียงช่วยเพิ่มประสิทธิภาพและความแม่นยำของการทำนายอาชญากรรมและจะส่งผลให้เป็นประโยชน์ต่อการนำไปประยุกต์ใช้เพื่อปรับกลยุทธ์การบังคับใช้กฎหมายมีประสิทธิผลมากขึ้น

Tutorial - Shodhguru Labs: Optimization And Hyperparameter Tuning For Neural Networks, Kaushik Roy

Publications

Neural networks have emerged as a powerful and versatile class of machine learning models, revolutionizing various fields with their ability to learn complex patterns and make accurate predictions. The performance of neural networks depends significantly on the appropriate choice of hyperparameters, which are critical factors governing their architecture, regularization, and optimization techniques. As the demand for high-performance neural networks grows across diverse applications, the need for efficient optimization and hyperparameter tuning methods becomes paramount. This paper presents a comprehensive exploration of optimization strategies and hyperparameter tuning techniques for neural networks. Neural networks have emerged as a powerful and versatile class …

Geometry-Aware Methodology For Coupling Mechanism Analysis And Radiation Mechanism Analysis, Xu Wang

Doctoral Dissertations

"Due to the wide-band spectrum of digital signals, a variety of noise generators, including microprocessors, liquid crystal displays, digital microphones, high-speed traces, and double-data-rate memory modules, are found in consumer electronics products. Noise generated by digital circuits can couple to the antennas and degrade their sensitivity. Additionally, it also couples to metallic housings or heatsinks and produces a significant amount of emission. In this research, methodologies are presented to analyze the coupling and radiation mechanism in electronic devices.

To investigate the coupling between the distributive geometries in the radio frequency interference (RFI) problem, a mesh-dependent partition-based coupling mechanism analysis method …

Characterization And Improvement Of Powder Properties Through Plasma Spheroidization For Use In The Laser Powder Bed Fusion Process, M. Hossein Sehhat

Doctoral Dissertations

"This research was conducted to characterize the powder properties using different characterization techniques to enhance the powder behavior in the Laser Powder Bed Fusion (LPBF) process. One of the shortcomings of LPBF process is the high cost of gas-atomized powder feedstock. To overcome this downside, the cheaper water-atomized powder was characterized and used in LPBF process, and its performance was compared with those of gas-atomized powder. Also, the plasma spheroidization process was used to improve the powder characteristics; if the spheroidized powder could function as well as the virgin powder, plasma spheroidization process could be applied to the used powders …

ผลของความโค้งของคลองรากฟันต่อความทนทานต่อแรงบิดของตะไบทางทันตกรรมแบบหมุน, ปุณยวุฒิ โตมีบุญ

Chulalongkorn University Theses and Dissertations (Chula ETD)

วิทยานิพนธ์นี้ศึกษาความทนทานต่อแรงบิดของตะไบทางทันตกรรมแบบหมุนที่ทำงานในคลองรากฟันที่มีความโค้งแบบต่าง ๆ ด้วยระเบียบวิธีไฟไนต์เอลิเมนต์และวิธีการทดลอง การศึกษานี้จำลองการรับแรงบิดของตะไบขนาด 0.25 และ 0.35 มิลลิเมตร ขณะทำงานในคลองรากฟันที่มีรัศมีความโค้ง 2, 5 และ 8 มิลลิเมตร และทำงานในคลองรากฟันที่มีมุมความโค้ง 30, 45 และ 60 องศา โดยมีตะไบตรงเป็นชุดควบคุม การศึกษาด้วยระเบียบวิธีไฟไนต์เอลิเมนต์พิจารณาแนวโน้มของค่าความเค้นและแรงบิดปฏิกิริยาที่ปลายตะไบภายใต้แรงบิดขนาดคงที่ การทดลองหาแรงบิดจากเครื่องกรอฟันที่ทำให้ตะไบเสียหายในขณะทำงานในคลองรากฟันที่มีความโค้งแบบต่าง ๆ ผลการศึกษาด้วยระเบียบวิธีไฟไนต์เอลิเมนต์พบว่าความเค้นและแรงบิดปฏิกิริยาที่เกิดขึ้นที่ปลายของตะไบที่ทำงานในคลองรากฟันโค้งมีค่าน้อยกว่าตะไบที่ทำงานในคลองรากฟันตรง ความเค้นและแรงบิดปฏิกิริยามีค่าน้อยลงเมื่อตะไบรับแรงบิดขณะวางตัวอยู่ในคลองรากฟันที่มีรัศมีความโค้งน้อยลง และมีค่าน้อยลงเมื่อรับแรงบิดขณะอยู่ในคลองรากฟันที่มีมุมความโค้งมากขึ้น ในส่วนของการทดลองพบว่าผลการทดลองสอดคล้องกับผลการศึกษาโดยระเบียบวิธีไฟไนต์เอลิเมนต์ ตะไบโค้งสามารถรับแรงบิดจากเครื่องกรอฟันได้มากกว่าตะไบตรง ความสามารถในการรับแรงบิดจากเครื่องกรอฟันของตะไบลดลงขณะตะไบทำงานในคลองรากฟันที่มีรัศมีความโค้งมากขึ้น แต่ตะไบสามารถรับแรงบิดได้มากขึ้นเมื่อทำงานในคลองรากฟันที่มีมุมความโค้งมากขึ้น นั่นคือ ผลการศึกษาจากระเบียบวิธีไฟไนต์เอลิเมนต์และวิธีการทดลองมีความสอดคล้องกัน ตะไบทางทันตกรรมแบบหมุนจะสามารถรับแรงบิดได้มากขึ้นเมื่อตะไบทำงานในคลองรากฟันที่มีความโค้ง