This dissertation is about a particular style of research. The philosophy of this style is that in order to scientifically understand deep learning, it is fruitful to investigate what happens when neural networks are trained on simple, mathematically well-defined tasks. Even though the training data is simple, the training algorithm can end up…

The field of Data Science has seen rapid growth over the past two decades, with a high demand for people with skills in data analytics, programming, statistics, and ability to visualize, predict from, and otherwise make sense of data. Alongside the rise of various artificial intelligence (AI) and machine learning (ML) applications, we have also…

Probabilistic machine learning increasingly informs critical decisions in medicine, economics, politics, and beyond. To aid the development of trust in these decisions, we develop a taxonomy delineating where trust in an analysis can break down: (1) in the translation of real-world goals to goals on a particular set of training data, (2) in the…

Immediacy is one of the necessary criteria to show strong evidence of treatment effect in single case experimental designs (SCEDs). With the exception of Natesan and Hedges (2017) no inferential statistical tool has been used to demonstrate or quantify it until now. We investigate and quantify immediacy by treating the change-points between the…

In our investigation of university students' knowledge about real numbers in relation to computer algebra systems (CAS) and how it could be developed in view of their future activity as teachers, we used a computer algorithm as a case to explore the relationship between CAS and the knowledge of real numbers as decimal representations. Our work was…

Language is one of the hallmarks of intelligence, demanding explanation in a theory of human cognition. However, language presents unique practical challenges for quantitative empirical research, making many linguistic theories difficult to test at naturalistic scales. Artificial neural network language models (LMs) provide a new tool for studying…

As we move further into the digital age, the acquisition of digital literacy (DL) and computational thinking (CT) skills is emerging internationally as an essential goal for students in contemporary school curricula. As the world becomes more uncertain and volatile due to impacts of artificial intelligence (AI), international unrest, climate…

Presents a case study that calculates the reduction of the time needed for a Web page to be fully downloaded and delivered to the user. Topics include bandwidth; HTTP; data transfer; algorithms for compression and decompression of Web pages; and reducing the size of hypertext data. (Author/LRW)

Discusses the development of an algorithm for an equitable distribution of state prevention funds to its substate jurisdiction based on the need for prevention services. Suggests that the final decision on the level of funding must be based on the selection of the schedule for resource allocation rather than the suggested amount or level of…

Discusses difficulties of producing computer assisted learning (CAL) software economically. Proposed as a solution, modular programing is illustrated by describing production of a matrix algebra package. Use of special CAL modules and facilities provided by subroutine libraries are cited, and development of a package to teach matrix theory is…

Presents model of algorithm design (activity in software development) based on analysis of protocols of two subjects designing three convex hull algorithms. Automation methods, methods for studying algorithm design, role of discovery in problem solving, and comparison of different designs of case study according to model are highlighted.…

The first of three articles that develop a framework for county-based prevention-resource-allocation algorithms based on the aggravated need for substance-abuse-prevention services estimated at the county level. Algorithm development is based on (1) statewide student drug survey, and (2) a set of social indicators routinely published by agencies…

Introduces a new methodology for studying the effects of term selection and postings overlap on retrieval, under the assumption that pooling results for several searchers or questions may obscure important relationships. Based on the findings of a case study, several tentative conclusions and hypotheses, are posited and suggestions for future…

Describes the history and development of the algorithm that was used to produce a union list of monographs for ALEPH (Automated Library Expandable Program Hebrew University), Israel's research library network. The literature on merging online records is reviewed, case studies are examined, and problems with matching and merging online…

A 14-year-old Brazilian boy in the fourth grade was given 170 problems orally, each asking for the sum of 2 2-digit numbers. Responses indicated he had invented his own algorithm for addition, similar to the school one, based largely on his experiences with money. (Contains 13 references.) (JAF)