arXiv subjects

Daniel Berleant

Publications and source records attributed to Daniel Berleant.

14 recordsLinked to original sources

arXiv2026

Distilled Rapid Embedding Transfer (DRET): Parameter-Efficient Biomedical Domain Adaptation via Priority-Based Embedding Transfer

Large domain-specific language models such as BioBERT and ClinicalBERT achieve strong performance on biomedical NLP tasks, but their computational demands make them impractical for many real-world deployments. General-purpose, parameter-efficient models such as DistilBERT are lightweight yet lack the domain knowledge required for specialized tasks such as PICO (Population, Intervention, Comparison, Outcome) classification. We introduce Distilled Rapid Embedding Transfer (DRET), a knowledge-transfer paradigm that injects biomedical domain knowledge from large specialized models into a smaller general-purpose model without retraining on the original specialized corpora. DRET is developed as an iterative family of strategies: a unified tokenizer-merge strategy (DRET 1.x), hybrid embedding averaging (DRET 2.0), and a priority-based embedding-transfer mechanism (DRET 3.x) that hierarchically selects embeddings from the most authoritative source models, further combined with embedding-layer freezing, differential learning rates, label propagation, and imbalance-aware loss functions (DRET 4.x). We evaluate DRET on token-level PICO classification using the EBM-NLP corpus under severe class imbalance, across a twelve-metric battery. DRET-enhanced DistilBERT (66M parameters) attains balanced accuracy, recall, and ROC-AUC competitive with, and on several class-wise metrics exceeding, models an order of magnitude larger, while retaining DistilBERT's efficiency. We further show that transfer occurs at the embedding level through cosine-similarity, semantic-shift, and t-SNE analyses. DRET offers a scalable, resource-efficient route to near-domain-expert performance for biomedical text mining, with direct application to automated systematic literature reviews and clinical decision support.

cs.CL↗

arXiv2025

Trend Extrapolation for Technology Forecasting: Leveraging LSTM Neural Networks for Trend Analysis of Space Exploration Vessels

Forecasting technological advancement in complex domains such as space exploration presents significant challenges due to the intricate interaction of technical, economic, and policy-related factors. The field of technology forecasting has long relied on quantitative trend extrapolation techniques, such as growth curves (e.g., Moore's law) and time series models, to project technological progress. To assess the current state of these methods, we conducted an updated systematic literature review (SLR) that incorporates recent advances. This review highlights a growing trend toward machine learning-based hybrid models. Motivated by this review, we developed a forecasting model that combines long short-term memory (LSTM) neural networks with an augmentation of Moore's law to predict spacecraft lifetimes. Operational lifetime is an important engineering characteristic of spacecraft and a potential proxy for technological progress in space exploration. Lifetimes were modeled as depending on launch date and additional predictors. Our modeling analysis introduces a novel advance in the recently introduced Start Time End Time Integration (STETI) approach. STETI addresses a critical right censoring problem known to bias lifetime analyses: the more recent the launch dates, the shorter the lifetimes of the spacecraft that have failed and can thus contribute lifetime data. Longer-lived spacecraft are still operating and therefore do not contribute data. This systematically distorts putative lifetime versus launch date curves by biasing lifetime estimates for recent launch dates downward. STETI mitigates this distortion by interconverting between expressing lifetimes as functions of launch time and modeling them as functions of failure time. The results provide insights relevant to space mission planning and policy decision-making.

cs.LG↗

arXiv2024

Quantitative Technology Forecasting: a Review of Trend Extrapolation Methods

Quantitative technology forecasting uses quantitative methods to understand and project technological changes. It is a broad field encompassing many different techniques and has been applied to a vast range of technologies. A widely used approach in this field is trend extrapolation. Based on the publications available to us, there has been little or no attempt made to systematically review the empirical evidence on quantitative trend extrapolation techniques. This study attempts to close this gap by conducting a systematic review of technology forecasting literature addressing the application of quantitative trend extrapolation techniques. We identified 25 studies relevant to the objective of this research and classified the techniques used in the studies into different categories, among which growth curves and time series methods were shown to remain popular over the past decade, while newer methods, such as machine learning-based hybrid models, have emerged in recent years. As more effort and evidence are needed to determine if hybrid models are superior to traditional methods, we expect to see a growing trend in the development and application of hybrid models to technology forecasting.

cs.AI↗

arXiv2023

ASI: Accuracy-Stability Index for Evaluating Deep Learning Models

In the context of deep learning research, where model introductions continually occur, the need for effective and efficient evaluation remains paramount. Existing methods often emphasize accuracy metrics, overlooking stability. To address this, the paper introduces the Accuracy-Stability Index (ASI), a quantitative measure incorporating both accuracy and stability for assessing deep learning models. Experimental results demonstrate the application of ASI, and a 3D surface model is presented for visualizing ASI, mean accuracy, and coefficient of variation. This paper addresses the important issue of quantitative benchmarking metrics for deep learning models, providing a new approach for accurately evaluating accuracy and stability of deep learning models. The paper concludes with discussions on potential weaknesses and outlines future research directions.

cs.LG↗

arXiv2023

Visual Question Answering (VQA) on Images with Superimposed Text

Superimposed text annotations have been under-investigated, yet are ubiquitous, useful and important, especially in medical images. Medical images also highlight the challenges posed by low resolution, noise and superimposed textual meta-information. Therefor we probed the impact of superimposing text onto medical images on VQA. Our results revealed that this textual meta-information can be added without severely degrading key measures of VQA performance. Our findings are significant because they validate the practice of superimposing text on images, even for medical images subjected to the VQA task using AI techniques. The work helps advance understanding of VQA in general and, in particular, in the domain of healthcare and medicine.

cs.CV↗

arXiv2022

Automating Systematic Literature Reviews with Natural Language Processing and Text Mining: a Systematic Literature Review

Objectives: An SLR is presented focusing on text mining based automation of SLR creation. The present review identifies the objectives of the automation studies and the aspects of those steps that were automated. In so doing, the various ML techniques used, challenges, limitations and scope of further research are explained. Methods: Accessible published literature studies that primarily focus on automation of study selection, study quality assessment, data extraction and data synthesis portions of SLR. Twenty-nine studies were analyzed. Results: This review identifies the objectives of the automation studies, steps within the study selection, study quality assessment, data extraction and data synthesis portions that were automated, the various ML techniques used, challenges, limitations and scope of further research. Discussion: We describe uses of NLP/TM techniques to support increased automation of systematic literature reviews. This area has attracted increase attention in the last decade due to significant gaps in the applicability of TM to automate steps in the SLR process. There are significant gaps in the application of TM and related automation techniques in the areas of data extraction, monitoring, quality assessment and data synthesis. There is thus a need for continued progress in this area, and this is expected to ultimately significantly facilitate the construction of systematic literature reviews.

cs.IR↗

arXiv2022

Discrete-Event Simulation in Healthcare Settings: a Review

We review and define the current state of the art as relating to discrete event simulation in healthcare-related systems. A review of published literature over the past five years (2017 - 2021) was conducted, building upon previously published work. PubMed and EBSCOhost were searched for journal articles on discrete event simulation in healthcare resulting in identification of 933 unique articles. Of these about half were excluded at the title/abstract level and 154 at the full text level, leaving 311 papers to analyze. These were categorized, then analyzed by category and collectively to identify publication volume over time, disease focus, activity levels by coun-try, software systems used, and sizes of healthcare unit under study. A total of 1196 articles were initially identified. This list was narrowed down to 311 for systematic review. Following the schema from prior systematic reviews, the articles fell into four broad categories: health care sys-tems operations (HCSO), disease progression modeling (DPM), screening modeling (SM), and health behavior modeling (HBM). We found that discrete event simulation in healthcare has con-tinued to increase year-over-year, as well as expand into diverse areas of the healthcare system. In addition, this study adds extra bibliometric dimensions to gain more insight into the details and nuances of how and where simulation is being used in healthcare.

cs.CY↗

arXiv2022

Discovering Limitations of Image Quality Assessments with Noised Deep Learning Image Sets

Image quality is important, and can affect overall performance in image processing and computer vision as well as for numerous other reasons. Image quality assessment (IQA) is consequently a vital task in different applications from aerial photography interpretation to object detection to medical image analysis. In previous research, the BRISQUE algorithm and the PSNR algorithm were evaluated with high resolution (atleast 512x384 pixels), but relatively small image sets (no more than 4,744 images). However, scientists have not evaluated IQA algorithms on low resolution (no more than 32x32 pixels), multi-perturbation, big image sets (for example, tleast 60,000 different images not counting their perturbations). This study explores these two IQA algorithms through experimental investigation. We first chose two deep learning image sets, CIFAR-10 and MNIST. Then, we added 68 perturbations that add noise to the images in specific sequences and noise intensities. In addition, we tracked the performance outputs of the two IQA algorithms with singly and multiply noised images. After quantitatively analyzing experimental results, we report the limitations of the two IQAs with these noised CIFAR-10 and MNIST image sets. We also explain three potential root causes for performance degradation. These findings point out weaknesses of the two IQA algorithms. The research results provide guidance to scientists and engineers developing accurate, robust IQA algorithms. All source codes, related image sets, and figures are shared on the website (https://github.com/caperock/imagequality) to support future scientific and industrial projects.

cs.CV↗

arXiv2022

Recent, rapid advancement in visual question answering architecture: a review

Understanding visual question answering is going to be crucial for numerous human activities. However, it presents major challenges at the heart of the artificial intelligence endeavor. This paper presents an update on the rapid advancements in visual question answering using images that have occurred in the last couple of years. Tremendous growth in research on improving visual question answering system architecture has been published recently, showing the importance of multimodal architectures. Several points on the benefits of visual question answering are mentioned in the review paper by Manmadhan et al. (2020), on which the present article builds, including subsequent updates in the field.

cs.CV↗

arXiv2022

Benchmarking Robustness of Deep Learning Classifiers Using Two-Factor Perturbation

Accuracies of deep learning (DL) classifiers are often unstable in that they may change significantly when retested on adversarial images, imperfect images, or perturbed images. This paper adds to the fundamental body of work on benchmarking the robustness of DL classifiers on defective images. To measure robust DL classifiers, previous research reported on single-factor corruption. We created comprehensive 69 benchmarking image sets, including a clean set, sets with single factor perturbations, and sets with two-factor perturbation conditions. The state-of-the-art two-factor perturbation includes (a) two digital perturbations (salt & pepper noise and Gaussian noise) applied in both sequences, and (b) one digital perturbation (salt & pepper noise) and a geometric perturbation (rotation) applied in both sequences. Previous research evaluating DL classifiers has often used top-1/top-5 accuracy. We innovate a new two-dimensional, statistical matrix to evaluating robustness of DL classifiers. Also, we introduce a new visualization tool, including minimum accuracy, maximum accuracy, mean accuracies, and coefficient of variation (CV), for benchmarking robustness of DL classifiers. Comparing with single factor corruption, we first report that using two-factor perturbed images improves both robustness and accuracy of DL classifiers. All source codes and related image sets are shared on the Website at http://cslinux.semo.edu/david/data to support future academic research and industry projects.

cs.CV↗

arXiv2021

Moore's law, Wright's law and the Countdown to Exponential Space

Technologies have often been observed to improve exponentially over time. In practice this often means identifying a constant known as the doubling time, describing the time period over which the technology roughly doubles in some measure of performance or of performance per dollar. Moore's law is, classically, the empirical observation that the number of electronic components that can be put on a chip doubles every 18 months to 2 years. Today it is frequently stated as the number of computations available per unit of cost. Generalized to the appropriate doubling time, it describes the rate of advancement in many technologies. A frequently noted competitor to Moore's law is known as Wright's law, which has aeronautical roots. Wright's law (also called power law, experience curve and Henderson's law) relates some quality of a manufactured unit (for Wright, airplanes) to the volume of units manufactured. The Wright's law equation expresses the idea that performance - price or a quality metric - improves according to a power of the number produced, or alternatively stated, improves by a constant percentage for every doubling of the total number produced. Does exploration of outer space conform to Moore's law or Wright's law-like behavior? Our results below are broadly consistent with these laws. This is true for many technologies. Although the two laws can make somewhat different predictions, Sahal found that they converge to the same predictions when manufacturing volume increases exponentially over time. When space exploration transitions into an independent commercial sector, as many people hope and expect, spacecraft technology will then likely enter an era of unambiguously exponential advancement.

q-fin.GN↗

arXiv2021

Benchmarking Robustness of Deep Learning Classifiers Using Two-Factor Perturbation

This paper adds to the fundamental body of work on benchmarking the robustness of deep learning (DL) classifiers. We innovate a new benchmarking methodology to evaluate robustness of DL classifiers. Also, we introduce a new four-quadrant statistical visualization tool, including minimum accuracy, maximum accuracy, mean accuracy, and coefficient of variation, for benchmarking robustness of DL classifiers. To measure robust DL classifiers, we created a comprehensive 69 benchmarking image set, including a clean set, sets with single factor perturbations, and sets with two-factor perturbation conditions. After collecting experimental results, we first report that using two-factor perturbed images improves both robustness and accuracy of DL classifiers. The two-factor perturbation includes (1) two digital perturbations (salt & pepper noise and Gaussian noise) applied in both sequences, and (2) one digital perturbation (salt & pepper noise) and a geometric perturbation (rotation) applied in both sequences. All source codes, related image sets, and preliminary data, figures are shared on a GitHub website to support future academic research and industry projects. The web resources locate at https://github.com/caperock/robustai

cs.LG↗

arXiv2019

Benchmarking Contemporary Deep Learning Hardware and Frameworks:A Survey of Qualitative Metrics

This paper surveys benchmarking principles, machine learning devices including GPUs, FPGAs, and ASICs, and deep learning software frameworks. It also reviews these technologies with respect to benchmarking from the perspectives of a 6-metric approach to frameworks and an 11-metric approach to hardware platforms. Because MLPerf is a benchmark organization working with industry and academia, and offering deep learning benchmarks that evaluate training and inference on deep learning hardware devices, the survey also mentions MLPerf benchmark results, benchmark metrics, datasets, deep learning frameworks and algorithms. We summarize seven benchmarking principles, differential characteristics of mainstream AI devices, and qualitative comparison of deep learning hardware and frameworks.

cs.DC↗

arXiv2004

Direct pivotal predictive inference

Without assuming any pdf for some measured parameter, we derive a predictive pdf for the outcome of a second measurement, given the outcome of the first measurement and two common assumptions about the noise. These are that (1) it is additive, and (2) it is of some known pdf. The argument is based on a Bayesian analysis of the noise when no pdf is provided for the value of the parameter. In this way we avoid assuming an ad-hoc prior. We clarify how this method of direct predictive inference is distinct from fiducial prediction. We specify the distinct flaw in the fiducial argument, and outline the importance of this development in the foundations of probability and statistics. Keywords: nonparametric predictive inference, direct pivotal argument, pivotal argument, fiducial argument, fiducial prediction, Bayesian inference, reference prior, reference class.

math.ST↗