Open Access. Powered by Scholars. Published by Universities.®

Digital Commons Network™

Open Access. Powered by Scholars. Published by Universities.®

Discipline
Institution
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 112141 - 112170 of 1113185

Full-Text Articles in Entire DC Network

Towards Multi-Modal Interpretable Video Understanding, Quang Sang Truong Dec 2023

Towards Multi-Modal Interpretable Video Understanding, Quang Sang Truong

Graduate Theses and Dissertations

This thesis introduces an innovative approach to video comprehension, which simulates human perceptual mechanisms and establishes a comprehensible and coherent narrative representation of video content. At the core of this approach lies the creation of a Visual-Linguistic (VL) feature for an interpretable video portrayal and an adaptive attention mechanism (AAM) aimed at concentrating solely on principal actors or pertinent objects while modeling their interconnections. Taking cues from the way humans disassemble scenes into visual and non-visual constituents, the proposed VL feature characterizes a scene via three distinct modalities: (i) a global visual environment, providing a broad contextual comprehension of the …

Go to article

Examining The Relationship Between Student Age And Modality Choice In A Community College’S New Student Orientation Program: A Preregistered Study To Examine The Assumption Of Self-Directed Learning, Amy L. Foree Dec 2023

Examining The Relationship Between Student Age And Modality Choice In A Community College’S New Student Orientation Program: A Preregistered Study To Examine The Assumption Of Self-Directed Learning, Amy L. Foree

Graduate Theses and Dissertations

Abstract Understanding and navigating college rules, resources, and expectations is a considerable barrier for students accessing and transitioning to college (Conley, 2007; Williams, 1996; Hooker & Brand, 2010; Ardoin, 2013; Sheppard, 2012). To improve students’ acquisition of pertinent, time-sensitive information, many institutions have implemented mandatory new student orientations (NSO). Orientations provide information such as academic policies and procedures, institutional expectations, campus resources, and financial aid assistance. They also allow students to meet faculty, staff, and other new students. By providing these resources, institutions acclimate students to their new environment. Although colleges realize that new student orientations support students’ success, they …

Go to article

The Alumni Base Holds The Institution In Their Heart: Alumni Decision Making And Communication Strategies During University Leadership Change, Brandy Cox Jackson Dec 2023

The Alumni Base Holds The Institution In Their Heart: Alumni Decision Making And Communication Strategies During University Leadership Change, Brandy Cox Jackson

Graduate Theses and Dissertations

The purpose of the research was to understand how alumni executives make decisions and formulate communication strategies during university leadership change. The research was based on the trends found by the American Council on Education that the tenure of the college president has been on the decline since 2006 and is currently 5.9 years. Multiple factors have been identified in the literature that have led to this decline. The research focused on the largest constituency of most universities, its alumni base, and provides context and a description of decision making and communication strategies for alumni executives navigating leadership change. The …

Go to article

Geospatial Analysis Of Weathering And Recession On Architectural Limestone At The 1891 Powerhouse/Icehouse In Eureka Springs, Arkansas, Stuart Bruce Wright Dec 2023

Geospatial Analysis Of Weathering And Recession On Architectural Limestone At The 1891 Powerhouse/Icehouse In Eureka Springs, Arkansas, Stuart Bruce Wright

Graduate Theses and Dissertations

The objective of this study of the architectural limestone of the 1891 Powerhouse/Icehouse in Eureka Springs, Arkansas was to analyze the environmental factors that influenced the deterioration and surface recession of this historical stone structure. The focus of this research was to examine and establish a baseline study that may assist in future monitoring and documentation of this historic limestone structure on stone weathering. A replicable methodology was specifically created toward this baseline research in hopes that it represented a solid baseline study. Observations and data analyses revealed the relationship between the architectural limestone building structure and the dominating recession …

Go to article

Exploring Success Factors Of Women Senior Leaders In The United States Paper Manufacturing Operations Industry: A Phenomenological Study, Ericka Eggleston Dec 2023

Exploring Success Factors Of Women Senior Leaders In The United States Paper Manufacturing Operations Industry: A Phenomenological Study, Ericka Eggleston

Graduate Theses and Dissertations

The purpose of this study is to understand how women describe their journey to becoming senior leaders within the operations sector of the Paper Manufacturing Industry. This study will provide insights about how women in the male-dominated Paper Manufacturing Industry (PMI) assumed leadership positions. This pre-registered study will offer the framework for this study to be carried out at the University of Arkansas. The researcher will apply a descriptive phenomenological approach that will use semi-structured questions to understand the strategies they used to navigate their career path. Snowball sampling will be used to select 5-15 participants who provide their perceptions …

Go to article

Tissue-Engineered Models For Investigating Nerve Infiltration In Fibrotic Microenvironments, Nikolas Ala-Kokko Dec 2023

Tissue-Engineered Models For Investigating Nerve Infiltration In Fibrotic Microenvironments, Nikolas Ala-Kokko

Graduate Theses and Dissertations

Neuron infiltration is involved in a multitude of diseased states and is critical in the regeneration of physiologic conditions. Modeling of innervation in three-dimensional, heterogenous, environments would allow early investigation in the crosstalk of neurons and native cells in normal and pathophysiologic microenvironments to gain an understanding of mechanistic functions. This would allow for greater insight into the role of metabolism, cell adhesion, microenvironmental alternations, cellular crosstalk, and signaling pathways that play a role in the promotion and prevention of neurite growth. In this thesis, with the construction of collagen-based hydrogels as the foundation for cell culture, we investigate cell …

Go to article

Health Beneficial Effects Of Black Raspberry, Blueberry, And Blackberry Volatile Extracts In Lung And Colon, Inah Gu Dec 2023

Health Beneficial Effects Of Black Raspberry, Blueberry, And Blackberry Volatile Extracts In Lung And Colon, Inah Gu

Graduate Theses and Dissertations

Berries are one of the most consumed fruits in our diet, and rich in dietary fiber, vitamins, minerals, polyphenols, and volatiles. Volatiles are responsible for the aroma of berries, and are widely used in food, perfumery, and cosmetic industries. However, the potential health beneficial effects of berry volatiles have not been extensively studied. Recently, we found that berry volatiles from cranberries, black and red raspberries, strawberries, blueberries, and blackberries have an anti-inflammatory effect on inflamed murine macrophage cells. Since inflammation can develop into many chronic diseases, and berry volatiles can be inhaled or consumed and pass through the lungs or …

Go to article

Effect Of Onion Peel Bioactive Compounds On Gut Microbiome Of Obese Adults, Saad Al-Din Sifat Dec 2023

Effect Of Onion Peel Bioactive Compounds On Gut Microbiome Of Obese Adults, Saad Al-Din Sifat

Graduate Theses and Dissertations

Obesity, a pervasive global health concern, involves a complex interplay between genetic factors, dietary patterns, and the gut microbiota. Onion peels are rich in polyphenols and known for their antioxidant properties. However, the effects of in-vitro digested onion peel extract has not been investigated thus far. This study therefore assesses the effects of digested bioactive compounds from onion peels, emphasizing gut microbiota shift and metabolites. Onion peel extract (OPE), digested onion peel extract (DOPE), and solid-phase-extracted digested onion peel extracts (SPE-DOPE) were analyzed for total phenolic content (TPC) and antioxidant activity. SPE-DOPE exhibited the highest TPC and antioxidant activity. Fecal …

Go to article

Solar Cooking Solutions For Haiti, Isaac Mergner Dec 2023

Solar Cooking Solutions For Haiti, Isaac Mergner

Graduate Theses and Dissertations

This study focuses on enhancing the efficiency of solar cooking through the use of solar cookers and heat retention boxes. The research involved testing various cooking vessels and lids in a solar cooking experiment, with a specific focus on the performance of different materials and their heat retention capabilities. The experiment aimed to increase the efficiency of solar cookers, which are popular in developing nations due to their relatively low cost. The findings indicate that the combination of a solar cooker and a heat retention box can significantly improve the cooking process, particularly in areas with limited access to cooking …

Go to article

Deep Learning For Photovoltaic Characterization, Adrian Manuel De Luis Garcia Dec 2023

Deep Learning For Photovoltaic Characterization, Adrian Manuel De Luis Garcia

Graduate Theses and Dissertations

This thesis introduces a novel approach to Photovoltaic (PV) installation segmentation by proposing a new architecture to understand and identify PV modules from overhead imagery. Pivotal to this concept is the creation of a new Transformer-based network, S3Former, which focuses on small object characterization and modelling intra- and inter- object differentiation inside an image. Accurate mapping of PV installations is pivotal for understanding their adoption and guiding energy policy decisions. Drawing insights from current Deep Learning methodologies for image segmentation and building upon State-of-the-Art (SOTA) techniques in solar cell mapping, this work puts forth S3Former with the following enhancements: 1. …

Go to article

Development Of Lignin-Based Copolymers And Polymer Blends, David Chem Dec 2023

Development Of Lignin-Based Copolymers And Polymer Blends, David Chem

Graduate Theses and Dissertations

Increasing industrial interest in reducing fossil fuels in virgin polymer production and incorporating bio-based polymeric materials has led researchers to explore the potential of utilizing plant-based polymers such as lignin. Chapter 1 of this thesis presents a comprehensive overview of copolymerization and blending techniques for lignin and synthetic polymers, emphasizing different types of modifications and copolymerization techniques as well as key parameters such as temperature, reaction time, pH level, and solution compositions. These valorization methods are directed towards the development of high-value lignin-based materials with properties well-suited for various commercial applications. Chapter 2 offers a fundamental exploration into the amination …

Go to article

Monitoring External Workloads And Countermovement Jump Performance Throughout A Preseason In Division 1 Collegiate Women’S Basketball Players, Michelle Van Dyke Dec 2023

Monitoring External Workloads And Countermovement Jump Performance Throughout A Preseason In Division 1 Collegiate Women’S Basketball Players, Michelle Van Dyke

Electronic Theses and Dissertations

Monitoring external workloads and countermovement jump performance may be useful for coaches. PURPOSE: The purpose of this study was to determine the effects of external load on player performance as measured by a CMJ and specific blood biomarkers throughout the preseason. METHODS:10 female division 1 basketball athletes had PlayerLoadTM (PL) monitored for all mandatory basketball training during six weeks of the preseason and CMJs were performed weekly. Blood biomarkers were collected before preseason and at the end of preseason. Data were analyzed via the Catapult Sport software (Openfield, Catapult, Innovations, Melbourne, VIC, Australia) to quantify all participant …

Go to article

Experimental And Theoretical Investigations Of Novel 2d Magnet Heterostructures And Magnetic Thin Films, Stephen Hofer Dec 2023

Experimental And Theoretical Investigations Of Novel 2d Magnet Heterostructures And Magnetic Thin Films, Stephen Hofer

Dissertations

Since the first usage of magnets as navigation tools, magnetism has been an integral part of human technology. As such, with increased demand for more sophisticated technologies, the need to understand magnetism and its applications has also increased. In modern times, the focus of new technologies is on scalability, or how things behave on the smallest scales. To this end it is necessary to understand the fundamental and practical behaviors of magnetic materials at this limit. Recent discoveries in the field of true 2D magnets and the field of topological matter demands investigation of the intersection between these two fields. …

Go to article

A New Approach To Synthetic Image Evaluation, Majid Memari Dec 2023

A New Approach To Synthetic Image Evaluation, Majid Memari

Dissertations

This study is dedicated to enhancing the effectiveness of Optical Character Recognition (OCR) systems, with a special emphasis on Arabic handwritten digit recognition. The choice to focus on Arabic handwritten digits is twofold: first, there has been relatively less research conducted in this area compared to its English counterparts; second, the recognition of Arabic handwritten digits presents more challenges due to the inherent similarities between different Arabic digits.OCR systems, engineered to decipher both printed and handwritten text, often face difficulties in accurately identifying low-quality or distorted handwritten text. The quality of the input image and the complexity of the text …

Go to article

Virtue-Driven Leadership: Powering Excellence In Organizations, Joseph Scherrer Dec 2023

Virtue-Driven Leadership: Powering Excellence In Organizations, Joseph Scherrer

Arts & Sciences Graduate Student Theses and Dissertations

In this thesis, I seek to answer the question “What makes a good leader?” I approach this question systematically, starting in Chapter 1 by asking “What is Leadership?” In attempting to formulate a response, I find that the concept is slipperier than it first appears and difficult to pin down. All the same, I construct a thematic, contextually pertinent definition that provides reasonable precision for the purposes of this study. In Chapter 2, I present a representative survey of the social-scientific academic literature in order to establish the prospect that a philosophy of virtuous leadership can be empirically validated in …

Go to article

Amorphous Boron Carbide-Amorphous Silicon Heterojunction Devices, Vojislav Medic Dec 2023

Amorphous Boron Carbide-Amorphous Silicon Heterojunction Devices, Vojislav Medic

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

This dissertation will show successful development and characterization of amorphous boron carbide-amorphous silicon heterojunction device with potential for neutron detection. The amorphous hydrogenated boron carbide (a-BC:H) has been extensively researched as a semiconductor for neutron voltaic device fabrication. Naturally occurring boron contains 19.8% of boron isotope B10 that has a high absorption cross section of thermal neutrons at lower energies, and boron carbide contains 14.7% of that B10 isotope. Therefore, as a semiconductor compound of boron a-BC:H has the ability to absorb radiation, generate charge carriers, and collect those carriers. Previous work on a-BC:H devices investigated the fabrication …

Go to article

Enhanced Privacy-Enabled Face Recognition Using Κ-Identity Optimization, Ryan Karl Dec 2023

Enhanced Privacy-Enabled Face Recognition Using Κ-Identity Optimization, Ryan Karl

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

Facial recognition is becoming more and more prevalent in the daily lives of the common person. Law enforcement utilizes facial recognition to find and track suspects. The newest smartphones have the ability to unlock using the user's face. Some door locks utilize facial recognition to allow correct users to enter restricted spaces. The list of applications that use facial recognition will only increase as hardware becomes more cost-effective and more computationally powerful. As this technology becomes more prevalent in our lives, it is important to understand and protect the data provided to these companies. Any data transmitted should be encrypted …

Go to article

An Investigation Of Match For Lossless Video Compression, Brittany Sullivan-Reicks Dec 2023

An Investigation Of Match For Lossless Video Compression, Brittany Sullivan-Reicks

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

A new lossless video compression technique, Match, is investigated. Match uses the similarity between the frames of a video or the slices of medical images to find a prediction for the current pixel. A portion of the previous frame is searched to find a matching context, which is the pixels surrounding the current pixel, within some distance centered on the current location. The best distance to use for each dataset is found experimentally. The matching context refers to the neighborhood of w, nw, n, and ne, where the pixel in the previous frame with the closest matching context becomes the …

Go to article

Low-Power, Event-Driven System On A Chip For Charge Pulse Processing Applications, Joseph A. Schmitz Dec 2023

Low-Power, Event-Driven System On A Chip For Charge Pulse Processing Applications, Joseph A. Schmitz

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

This dissertation presents an electronic architecture and methodology capable of processing charge pulses generated by a range of sensors, including radiation detectors and tactile synthetic skin. These sensors output a charge signal proportional to the input stimulus, which is processed electronically in both the analog and digital domains. The presented work implements this functionality using an event-driven methodology, which greatly reduces power consumption compared to standard implementations. This enables new application areas that require a long operating time or compact physical dimensions, which would not otherwise be possible. The architecture is designed, fabricated, and tested in the aforementioned applications to …

Go to article

Iron Promotes Glycolysis To Drive Colon Tumorigenesis., Zhaoli Liu, Luke Villareal, Lavanya Goodla, Hyeoncheol Kim, Daniel M. Falcon, Mohammad Haneef, David R. Martin, Li Zhang, Ho-Joon Lee, Daniel Kremer, Costas A. Lyssiotis, Yatrik M. Shah, Henry C. Lin, Hui-Kuan Lin, Xiang Xue Dec 2023

Iron Promotes Glycolysis To Drive Colon Tumorigenesis., Zhaoli Liu, Luke Villareal, Lavanya Goodla, Hyeoncheol Kim, Daniel M. Falcon, Mohammad Haneef, David R. Martin, Li Zhang, Ho-Joon Lee, Daniel Kremer, Costas A. Lyssiotis, Yatrik M. Shah, Henry C. Lin, Hui-Kuan Lin, Xiang Xue

Pathology Research and Scholarship

Colorectal cancer (CRC) is the third most common cancer and is also the third leading cause of cancer-related death in the USA. Understanding the mechanisms of growth and progression of CRC is essential to improve treatment. Macronutrients such as glucose are energy source for a cell. Many tumor cells exhibit increased aerobic glycolysis. Increased tissue micronutrient iron levels in both mice and humans are also associated with increased colon tumorigenesis. However, if iron drives colon carcinogenesis via affecting glucose metabolism is still not clear. Here we found the intracellular glucose levels in tumor colonoids were significantly increased after iron treatment.

Go to article

Post-Wildfire Export Regimes Of Solutes Along The Gallinas-Pecos River-Santa Rosa Fluvial Network, Asmita Kaphle Dec 2023

Post-Wildfire Export Regimes Of Solutes Along The Gallinas-Pecos River-Santa Rosa Fluvial Network, Asmita Kaphle

Civil Engineering ETDs

Wildfires significantly contribute to water pollution when ash and sediments, including nutrients and metals, are released into fluvial systems, impacting aquatic ecosystems and surrounding communities. This study focuses on the impacts of the Hermits Peak-Calf Canyon Fire on the Gallinas and Pecos watersheds. It employs five monitoring stations along a 190km fluvial network to assess solute export regimes post-fire, where grab samples were collected for nutrient and metal analyses for a period of more than a year. The metrics used for characterizing mobilization, chemostatic, or dilution patterns indicated mobilization for most solutes at all sites upstream of Santa Rosa Lake. …

Go to article

Extremum Seeking Control Algorithms For Extremely High Frequency Antenna System Pointing, Scott Shore Dec 2023

Extremum Seeking Control Algorithms For Extremely High Frequency Antenna System Pointing, Scott Shore

Mechanical Engineering ETDs

The research examines the application of extremum seeking control (ESC) algorithms to ground station antenna pointing for extremely high frequency (EHF) communication systems. ESC algorithms search for local maxima or minima by locating where an objective function gradient goes to zero. With wireless communication expanding into higher frequencies, the ground station pointing requirement is increasing. ESC presents a method which utilizes available equipment and information to perform ground station pointing. Additionally, ESC algorithms do not rely on assumptions or approximations needed for other techniques. The dissertation demonstrates ESC algorithm feasibility for the ground station pointing problem, benchmarks the ESC algorithms …

Go to article

Research Performance Analysis Of Top Six Heis In Wb Based On Nirf-2021 Ranking: An Evaluation-Based Study, Debdas Mondal Dec 2023

Research Performance Analysis Of Top Six Heis In Wb Based On Nirf-2021 Ranking: An Evaluation-Based Study, Debdas Mondal

Library Philosophy and Practice (e-journal)

The Indian higher education system is the third largest education system in the world, however, no Indian universities have been found in the list of top 100 universities in the world. The Government of India has taken several initiatives to improve the quality of education in India and its ranking position so that Indian universities improve their quality and provide the best quality education and research. NIRF is one such initiative. As per NIRF, all the HEIs are evaluated on the basis of certain stipulated performance of parameters score, such as ’Quantitative Research (QNR)’, ’Qualitative Research (QLR)’, ’Students and Faculty …

Go to article

Llm-Adapters: An Adapter Family For Parameter-Efficient Fine-Tuning Of Large Language Models, Zhiqiang Hu, Lei Wang, Yihuai Lan, Wanyu Xu, Ee-Peng Lim, Lidong Bing, Xing Xu, Soujanya Poria, Roy Ka-Wei Lee Dec 2023

Llm-Adapters: An Adapter Family For Parameter-Efficient Fine-Tuning Of Large Language Models, Zhiqiang Hu, Lei Wang, Yihuai Lan, Wanyu Xu, Ee-Peng Lim, Lidong Bing, Xing Xu, Soujanya Poria, Roy Ka-Wei Lee

Research Collection School Of Computing and Information Systems

The success of large language models (LLMs), like GPT-4 and ChatGPT, has led to the development of numerous cost-effective and accessible alternatives that are created by finetuning open-access LLMs with task-specific data (e.g., ChatDoctor) or instruction data (e.g., Alpaca). Among the various fine-tuning methods, adapter-based parameter-efficient fine-tuning (PEFT) is undoubtedly one of the most attractive topics, as it only requires fine-tuning a few external parameters instead of the entire LLMs while achieving comparable or even better performance. To enable further research on PEFT methods of LLMs, this paper presents LLMAdapters, an easy-to-use framework that integrates various adapters into LLMs and …

Go to article

Rome: Evaluating Pre-Trained Vision-Language Models On Reasoning Beyond Visual Common Sense, Kankan Zhou, Eason Lai, Au Wei Bin Yeong, Kyriakos Mouratidis, Jing Jiang Dec 2023

Rome: Evaluating Pre-Trained Vision-Language Models On Reasoning Beyond Visual Common Sense, Kankan Zhou, Eason Lai, Au Wei Bin Yeong, Kyriakos Mouratidis, Jing Jiang

Research Collection School Of Computing and Information Systems

Humans possess a strong capability for reasoning beyond common sense. For example, given an unconventional image of a goldfish laying on the table next to an empty fishbowl, a human would effortlessly determine that the fish is not inside the fishbowl. The case, however, may be different for a vision-language model, whose reasoning could gravitate towards the common scenario that the fish is inside the bowl, despite the visual input. In this paper, we introduce a novel probing dataset named ROME (reasoning beyond commonsense knowledge) to evaluate whether the state-of-the-art pre-trained vision-language models have the reasoning capability to correctly interpret …

Go to article

Examining The Inter-Consistency Of Large Language Models: An In-Depth Analysis Via Debate, Kai Xiong, Xiao Ding, Yixin Cao, Ting Liu, Bing Qin Dec 2023

Examining The Inter-Consistency Of Large Language Models: An In-Depth Analysis Via Debate, Kai Xiong, Xiao Ding, Yixin Cao, Ting Liu, Bing Qin

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) have shown impressive capabilities in various applications, but they still face various inconsistency issues. Existing works primarily focus on the inconsistency issues within a single LLM, while we complementarily explore the inter-consistency among multiple LLMs for collaboration. To examine whether LLMs can collaborate effectively to achieve a consensus for a shared goal, we focus on commonsense reasoning, and introduce a formal debate framework (FORD) to conduct a three-stage debate among LLMs with real-world scenarios alignment: fair debate, mismatched debate, and roundtable debate. Through extensive experiments on various datasets, LLMs can effectively collaborate to reach a consensus …

Go to article

Learning To Search Feasible And Infeasible Regions Of Routing Problems With Flexible Neural K-Opt, Yining Ma, Zhiguang Cao, Yew Meng Chee Dec 2023

Learning To Search Feasible And Infeasible Regions Of Routing Problems With Flexible Neural K-Opt, Yining Ma, Zhiguang Cao, Yew Meng Chee

Research Collection School Of Computing and Information Systems

In this paper, we present Neural k-Opt (NeuOpt), a novel learning-to-search (L2S) solver for routing problems. It learns to perform flexible k-opt exchanges based on a tailored action factorization method and a customized recurrent dual-stream decoder. As a pioneering work to circumvent the pure feasibility masking scheme and enable the autonomous exploration of both feasible and infeasible regions, we then propose the Guided Infeasible Region Exploration (GIRE) scheme, which supplements the NeuOpt policy network with feasibility-related features and leverages reward shaping to steer reinforcement learning more effectively. Besides, we further equip NeuOpt with dynamic data augmentations during inference for more …

Go to article

Exgen: Ready-To-Use Exercise Generation In Introductory Programming Courses, Nguyen Binh Duong Ta, Hua Gia Phuc Nguyen, Gottipati Swapna Dec 2023

Exgen: Ready-To-Use Exercise Generation In Introductory Programming Courses, Nguyen Binh Duong Ta, Hua Gia Phuc Nguyen, Gottipati Swapna

Research Collection School Of Computing and Information Systems

In introductory programming courses, students as novice programmers would benefit from doing frequent practices set at a difficulty level and concept suitable for their skills and knowledge. However, setting many good programming exercises for individual learners is very time-consuming for instructors. In this work, we propose an automated exercise generation system, named ExGen, which leverages recent advances in pre-trained large language models (LLMs) to automatically create customized and ready-to-use programming exercises for individual students ondemand. The system integrates seamlessly with Visual Studio Code, a popular development environment for computing students and software engineers. ExGen effectively does the following: 1) maintaining …

Go to article

Spatial-Temporal Episodic Memory Modeling For Adls: Encoding, Retrieval, And Prediction, Xinjing Song, Di Wang, Chai Quek, Ah-Hwee Tan, Yanjiang Wang Dec 2023

Spatial-Temporal Episodic Memory Modeling For Adls: Encoding, Retrieval, And Prediction, Xinjing Song, Di Wang, Chai Quek, Ah-Hwee Tan, Yanjiang Wang

Research Collection School Of Computing and Information Systems

Activities of daily living (ADLs) relate to people’s daily self-care activities, which reflect their living habits and lifestyle. A prior study presented a neural network model called STADLART for ADL routine learning. In this paper, we propose a cognitive model named Spatial-Temporal Episodic Memory for ADL (STEM-ADL), which extends STADLART to encode event sequences in the form of distributed episodic memory patterns. Specifically, STEM-ADL encodes each ADL and its associated contextual information as an event pattern and encodes all events in a day as an episode pattern. By explicitly encoding the temporal characteristics of events as activity gradient patterns, STEM-ADL …

Go to article

Generalized Logit Adjustment: Calibrating Fine-Tuned Models By Removing Label Bias In Foundation Models, Beier Zhu, Kaihua Tang, Qianru Sun, Hanwang Zhang Dec 2023

Generalized Logit Adjustment: Calibrating Fine-Tuned Models By Removing Label Bias In Foundation Models, Beier Zhu, Kaihua Tang, Qianru Sun, Hanwang Zhang

Research Collection School Of Computing and Information Systems

Foundation models like CLIP allow zero-shot transfer on various tasks without additional training data. Yet, the zero-shot performance is less competitive than a fully supervised one. Thus, to enhance the performance, fine-tuning and ensembling are also commonly adopted to better fit the downstream tasks. However, we argue that such prior work has overlooked the inherent biases in foundation models. Due to the highly imbalanced Web-scale training set, these foundation models are inevitably skewed toward frequent semantics, and thus the subsequent fine-tuning or ensembling is still biased. In this study, we systematically examine the biases in foundation models and demonstrate the …

Go to article

First Prev Next Last