Todd W. Neller

Enhancing undergraduate AI courses through machine learning projects

It is generally recognized that an undergraduate introductory artificial intelligence course is challenging to teach. This is, in part, due to the diverse and seemingly disconnected core topics that are typically covered. The paper presents work funded by the National Science Foundation to address this problem and to enhance the student learning experience in the course. Our work involves the development of an adaptable framework for the presentation of core AI topics through a unifying theme of machine learning. A suite of hands-on semester-long projects are developed, each involving the design and implementation of a learning system that enhances a commonly-deployed application. The projects use machine learning as a unifying theme to tie together the core AI topics. In this paper, we will first provide an overview of our model and the projects being developed and will then present in some detail our experiences with one of the projects, Web User Profiling, which we have used in our AI class

Pedagogical Possibilities for the N-Puzzle Problem

In this paper we present work on a project funded by the National Science Foundation with a goal of unifying the artificial intelligence (AI) course around the theme of machine learning. Our work involves the development and testing of an adaptable framework for the presentation of core AI topics that emphasizes the relationship between AI and computer science. Several hands-on laboratory projects that can be closely integrated into an introductory AI course have been developed. We present an overview of one of the projects and describe the associated curricular materials that have been developed. The project uses machine learning as a theme to unify core AI topics in the context of the N-puzzle game. Games provide a rich framework to introduce students to search fundamentals and other core AI concepts. The paper presents several pedagogical possibilities for the N-puzzle game, the rich challenge it offers, and summarizes our experiences using it

MLeXAI: A Project-Based Application-Oriented Model

Our approach to teaching introductory artificial intelligence (AI) unifies its diverse core topics through a theme of machine learning, and emphasizes how AI relates more broadly with computer science. Our work, funded by a grant from the National Science Foundation, involves the development, implementation, and testing of a suite of projects that can be closely integrated into a one-term AI course. Each project involves the development of a machine learning system in a specific application. These projects have been used in six different offerings over a three-year period at three different types of institutions. While we have presented a sample of the projects as well as limited preliminary experiences in other venues, this article presents the first assessment of our work over an extended period of three years. Results of assessment show that the projects were well received by the students. By using projects involving real-world applications we provided additional motivation for students. While illustrating core concepts, the projects introduced students to an important area in computer science, machine learning, thus motivating further study.

Implementing the intelligent systems knowledge units of computing curricula 2001

Computing curricula 2001 (CC-2001) presents a set of curricular recommendations for undergraduate computer science programs. CC-2001 presents a computer science body of knowledge and identifies a list of core topics/units within each component body of knowledge that a computer science program should require. While some of these core units span hours that warrant or are equivalent to a full course, the core units for other areas are significantly less. This paper presents our experiences with integrating intelligent systems (IS) core units of CC-2001 into the undergraduate curriculum through the more traditional core courses such as discrete mathematics, data structures, and algorithms, thus eliminating the need to require a full course in the area in departments with various constraints that prevent this from being possible.

Teaching AI through machine learning projects

An introductory Artificial Intelligence (AI) course provides students with basic knowledge of the theory and practice of AI as a discipline concerned with the methodology and technology for solving problems that are difficult to solve by other means. It is generally recognized that an introductory Artificial Intelligence course is challenging to teach. This is, in part, due to the diverse and seemingly disconnected core AI topics that are typically covered. Recently, work has been done to address the diversity of topics covered in the course and to create a theme-based approach. Russell and Norvig present an agent-centered approach [9]. Others have been working to integrate Robotics into the AI course [1, 2, 3].We present work on a project funded by the National Science Foundation with a goal of unifying the artificial intelligence course around the theme of machine learning. This involves the development and testing of an adaptable framework for the presentation of core AI topics that emphasizes the relationship between AI and computer science. Machine learning is inherently connected with the AI core topics and provides methodology and technology to enhance real-world applications within many of these topics. Machine learning also provides a bridge between AI technology and modern software engineering. In his article, Mitchell discusses the increasingly important role that machine learning plays in the software world and identifies three important areas: data mining, difficult-to-program applications, and customized software applications [6].We have developed a suite of adaptable, hands-on laboratory projects that can be closely integrated into the introductory AI course. Each project involves the design and implementation of a learning system which will enhance a particular commonly-deployed application. The goal is to enhance the student learning experience in the introductory artificial intelligence course by (1) introducing machine learning elements into the AI course, (2) implementing a set of unifying machine learning laboratory projects to tie together the core AI topics, and (3) developing, applying, and testing an adaptable framework for the presentation of core AI topics which emphasizes the important relationship between AI and computer science in general, and software development in particular. Details on this project as well as samples of course materials developed are published in [4, 5, 7, 8] and are available at the project website at http://uhaweb.hartford.edu/compsci/ccli.We present an overview of our work along with a detailed presentation of one of these projects and how it meets our goals.The project involves the development of a learning system for web document classification. Students investigate the process of classifying hypertext documents, called tagging, and apply machine learning techniques and data mining tools for automatic tagging. Our experiences using the projects are also presented.

Approximating Optimal Dudo Play with Fixed-Strategy Iteration Counterfactual Regret Minimization

Using the bluffing dice game Dudo as a challenge domain, we abstract information sets by an imperfect recall of actions. Even with such abstraction, the standard Counterfactual Regret Minimization (CFR) algorithm proves impractical for Dudo, since the number of recursive visits to the same abstracted information sets increase exponentially with the depth of the game graph. By holding strategies fixed across each training iteration, we show how CFR training iterations may be transformed from an exponential-time recursive algorithm into a polynomial-time dynamic-programming algorithm, making computation of an approximate Nash equilibrium for the full 2-player game of Dudo possible for the first time.

Information-Based Optimization Approaches to Dynamical System Safety Verification

Given a heuristic estimate of the relative safety of a hybrid dynamical system trajectory, we transform the initial safety problem for dynamical systems into a global optimization problem. We introduce MLLO-IQ and MLLO-RIQ, two new information-based optimization algorithms. After demonstrating their strengths and weaknesses, we describe the class of problems for which different optimization methods are best-suited.

Educational advances in artificial intelligence

In 2010 a new annual symposium on Educational Advances in Artificial Intelligence (EAAI) was launched as part of the AAAI annual meeting. The event was held in cooperation with ACM SIGCSE and has many similar goals related to broadening and disseminating work in computer science education. EAAI has a particular focus, however, as the event is specific to educational work in Artificial Intelligence and collocated with a major research conference (AAAI) to promote more interaction between researchers and educators in that domain. This panel seeks to introduce participants to EAAI as a way of fostering more interaction between educational communities in computing. Specifically, the panel will discuss the goals of EAAI, provide an overview of the kinds of work presented at the symposium, and identify potential synergies between that EAAI and SIGCSE as a way of better linking the two communities going forward.

Rook jumping maze design considerations

We define the Rook Jumping Maze, provide historical perspective, and describe a generation method for such mazes. When applying stochastic local search algorithms to maze design, most creative effort concerns the definition of an objective function that rates maze quality. We define and discuss several maze features to consider in such a function definition. Finally, we share our preferred design choices, make design process observations, and note the applicability of these techniques to variations of the Rook Jumping Maze.

Artificial Intelligence Education: Editorial Introduction

This issue of AI Magazine include five articles covering subjects of current concern to the AI education community. This editorial introduces those five articles.