The thesis focuses on model-based approximation methods for reinforcement
learning with large scale applications such as combinatorial optimization problems.
First, the thesis proposes two new model-based methods to stablize the
value–function approximation for reinforcement learning. The first one is the
BFBP algorithm, a batch-like reinforcement learning process which iterates between...
The code reuse problem is a common software engineering problem in scientific computing. As a prevailing programming language in many scientific fields, Fortran does not provide support to address this problem. One particular reason is that Fortran lacks the support for generic programming. By applying program-generation techniques, we developed two...
Software maintenance accounts for a large portion of the software development cost, particularly the process of updating programs either to adapt for requirement change or to enhance design or efficiency. Currently, program updates are generally performed manually by programmers using text editors. This is an unreliable
method because syntax and...
Spreadsheets are among the most widely used end-user programming systems. Unfortunately, there is a high incidence of errors in end-user spreadsheets, and some of these errors have high impact. In this dissertation, we describe techniques we have developed to help end users develop safer spreadsheets. As part of our research,...
Until recently, research has not considered whether the design of end-user programming environments, such as spreadsheets, multimedia authoring languages, and CAD systems, affects males and females differently. As a result, we began investigating how the two genders are impacted by end-user programming software and whether attention to gender differences is...
Accessing information on the Web has become ingrained into our daily lives, and we seek information from many different sources, including conference and journal publications, personal web pages, and others. Increasingly, web-based information retrieval systems such as web-based search engines, library on-line catalog systems, and subscription-based federated search systems are...
An n-bit Gray code is an ordered set of all 2n binary strings of length n. The
special property of this listing is that Hamming distance between consecutive vectors
is exactly 1. If the last and first codeword also have a Hamming distance 1 then the
code is said to...
The goal of many machine learning problems can be formalized as the creation of a function that can properly classify an input vector, given a set of examples of that function. While this formalism has produced a number of success stories, there are notable situations in which it fails. One...
Image feature detection and matching are two critical processes for many computer vision tasks. Currently, intensity-based local interest region detectors and local feature-based matching methods are used widely in computer vision applications. But in some applications, such as biological object recognition tasks, within-class changes in pose, lighting, color, and texture...
Building intelligent computer assistants has been a long-cherished goal of AI. Many intelligent assistant systems were built and fine-tuned to specific application domains. In this work, we develop a general model of assistance that combines three powerful ideas: decision theory, hierarchical task models and probabilistic relational languages. We use the...