Artificial Intelligence (AI) Course Group Status Report

1. Summary

The field of AI continues to evolve at a rapid pace, fueled by the ancient dream of the mankind to make a machine that behaves like a man, as well as by the research and development taking place in industry at companies such as Microsoft, IBM, AT&T, Lucent, GE, just to name a few. In addition, AI has become indispensable in a variety of application areas because of advances in computer vision, speech and language processing, inference engines, knowledge-based systems, and robotics. AI has also become a main bridge between computer science and many other disciplines, such as other engineering disciplines, cognitive science, linguistics, psychology, neuroscience, math/physics/statistics, speech and hearing sciences, and medicine.

We have evolved the AI curriculum continuously to keep pace with the development of the field, the interests of students, and the demands of employers. We currently have nine regular courses in AI, many of which are cross-listed with other departments, reflecting the interdisciplinary nature of the field. The 630/730 sequence is the core introduction to AI. The AI faculty also offer research seminars that cover recent developments in computer vision, neural networks, speech processing, knowledge systems, and cognitive science. Developing new courses on these topics is a priority of the AI faculty.

The AI curriculum satisfies many of the CSE and ABET objectives. On the application side, AI is an area in which students can exercise the principles of software design, databases, algorithms, data structures and programming languages which they have learned in the foundation courses. The AI courses rely heavily on the application of various principles from mathematics, physics, statistics, neuroscience, and psychology. Students who study AI in the Department receive a good grounding in the area, are usually ready for graduate studies in AI and cognitive science, and are very employable. AI also draws from an interdisciplinary background, and several of our courses offer opportunities for working with interdisciplinary teams.

2. Detailed Analysis

Section 2.1 describes the individual courses in the group. Section 2.2 explains how the group is related to the rest of the CSE program. Section 2.3 explains how the group helps meet a range of CSE and ABET objectives. Section 2.4 provides information on the feedback we have received from students, recruiters, etc., about the courses in the group. Section 2.5 summarizes changes to the course group since the last report. Section 2.6 summarizes continuing concerns and future changes under discussion.

2.1 Summary of the courses

CSE 630: Survey of Artificial Intelligence I: Basic Techniques  is a survey of the basic concepts and techniques of problem solving paradigms and knowledge representation schemes in Artificial Intelligence. This is the entry-level AI course, and is a prerequisite for many subsequent AI courses.

CSE 634: Computer Vision for Human Computer Interaction  is a course in computer vision algorithms for use in human-computer interactive systems. Topics include image formation, image features, segmentation, shape analysis, object tracking, motion calculation, and applications. This is a newly approved course (634) that was successfully piloted three times under CSE 694I; Autumn 2005 should be the first offering of the permanent course.

CSE 730: Survey of Artificial Intelligence II: Advanced Topics is a survey of advanced concepts, techniques, and applications of artificial intelligence, including perception, learning, reasoning under uncertainty, natural language processing, and speech recognition. This course continues the introduction of artificial intelligence in CSE 630.

CSE 731: Knowledge-based Systems is a survey of theory and practice of expert systems and knowledge-based systems, and makes use of current knowledge-based systems software tools. This is a BSCSE capstone design course.

CSE 732: Computational Linguistics is an exploration of the computational processing of natural language; syntactic, semantic, and pragmatic processing techniques are applied to understanding and generating written and spoken language. This course has been recently redesigned and has been offered twice in its new format.

CSE 735: Methods in Pattern Recognition is a survey of pattern recognition and machine learning techniques. This course has not been offered recently, but the AI faculty hope to resurrect it.

CSE 737: Proseminar in Cognitive Science is an in-depth examination of the interdisciplinary field of cognitive science, and it emphasizes fundamental issues of each discipline and provides illustrations of representative research being conducted at OSU. This is the second course in cognitive science, subsequent to CSE 612. The course is cross-listed with ISE, Linguistics, Philosophy, Psychology, and Speech and Hearing Science.

CSE 739: Knowledge-Based Systems in Engineering is an application of knowledge-based system principles to engineering problems, including practical knowledge engineering, techniques for problem assessment, and implementation. This course is cross-listed with Chemical, Mechanical, and Nuclear Engineering Departments.

CSE 776: Hardware/Software Interface Design Project introduces principles and applications of hardware and software design: design, programming, testing, and evaluation of an autonomous mobile robot system. This is a BSCSE capstone design course; it has not been offered recently.

CSE 779: Introduction to Neural Networks is a survey of fundamental methods and techniques of the field of artificial neural networks. Single-layer and multi-layer feedforward networks; Associative memory models; Recurrent and statistical networks; Supervised and unsupervised learning rules; Self-organization networks; Applications to signal processing, pattern recognition/generation, and optimization problems. This course is cross-listed with Electrical Engineering.

CSE 794L: Foundations of Spoken Language Processing is a new pilot course that covers automatic speech recognition and text-to-speech technologies. This course was first piloted in Winter 2005 and will be likely piloted again in Winter 2006.

2.2 Evaluation of courses

CSE 630 lists the following learning objectives:

• Master basic search techniques for problem-solving, including systematic blind search, heuristically-guided search, and optimal search.
• Be familiar with game tree search methods and the requirements for expert-level game play.
• Be familiar with using logic and proof as a basis for knowledge representation and automated reasoning.
• Be familiar with using semantic nets and frame systems as knowledge-representation formalisms.
• Be exposed to problems in common sense reasoning and language understanding.
• Be exposed to integrated AI architectures as a platform for building AI systems.
• Be exposed to machine learning techniques and the kinds of problem they solve.

• Master basic search techniques for problem-solving, including systematic blind search, heuristically-guided search, and optimal search.
• Master use of logic and proof as a basis for knowledge representation and automated reasoning.
• Be familiar with agent-based approaches to problem formulation.
• Be familiar with game tree search methods and the requirements for expert-level game play.
• Be exposed to problems in common sense reasoning and language understanding.
• Be exposed to machine learning techniques and the kinds of problem they solve.

CSE 731 lists the following learning objectives:

• Master task-level analysis and problem solving methods for classification problems.
• Master task-level analysis and problem solving methods for configuration (design) problems.
• Be familiar with the analysis and methods of diagnosis problems.
• Be familiar with methods for representing and reasoning with uncertain knowledge.
• Be familiar with task-level architectures.
• Be familiar with at least one modern programming tool for knowledge system development.

CSE 732 lists the following learning objectives:

• Master some current techniques for language processing.
• Master the application of symbolic and probabilistic data to language processing tasks.
• Be familiar with building interoperable components for a small language processing agent.
• Be familiar with the impact of context on language processing.

2.3 Relation of the courses to the rest of the CSE program

CSE 630: Survey of Artificial Intelligence I: Basic Techniques. Prerequisites for CSE 630 are CSE 222 (Development of Software Components) and Math 366 (Discrete Mathematical Structures I). CSE 630 is a prerequisite for CSE 634, 730, 731, 732, 735, 739, 779, and 794L.

CSE 634: Computer Vision for Human-Computer Interaction. Prerequisites for CSE 634 are CSE 630 (Survey of Artificial Intelligence I: Basic Techniques) or ECE 352 (Systems II); Math 568 (Introductory Linear Algebra I) or Math 571 (Linear Algebra for Applications I); or permission of instructor.

CSE 730: Survey of Artificial Intelligence II: Advanced Topics. Prerequisite for CSE 730 is CSE 630 (Survey of Artificial Intelligence I: Basic Techniques). CSE 730 is a prerequisite for CSE 794L.

CSE 731: Knowledge-based Systems. Prerequisite for CSE 731 is CSE 630 (Survey of Artificial Intelligence I: Basic Techniques), CSE 560 (Elements of Computer Systems Programming), and CSE 601 (Social and Ethical Issues in Computing), or permission of instructor.

CSE 732: Computational Linguistics. Prerequisites for CSE 732 are CSE 630 (Survey of Artificial Intelligence I: Basic Techniques); CSE 625 (Introduction to Automata and Formal Languages) or Linguistics 684.01 (Introduction to Computational Linguistics I); and Linguistics 201 (Introduction to Language in the Humanities), or permission of instructor.

CSE 735: Methods for Pattern Recognition. Prerequisites for CSE 735 are CSE 630 (Survey of Artificial Intelligence I: Basic Techniques) or permission of instructor.

CSE 737: Proseminar in Cognitive Science. Prerequisite for CSE 737 is CSE 612 (Introduction to Cognitive Science) or permission of instructor.

CSE 739: Knowledge-Based Systems in Engineering. Prerequisite for CSE 739 is CSE 630 (Survey of Artificial Intelligence I: Basic Technique) or permission of instructor.

CSE 776: Hardware/Software Interface Design Project. Prerequisites for CSE 776 are CSE 459.21 (Programming in C), CSE 660 (Introduction to Operating Systems); ECE 567 (Microprocessor Laboratory I) or ECE 329 (Electronic Devices and Control Laboratory).

CSE 779: Introduction to Neural Networks. Prerequisite for CSE 779 is CSE 630 (Survey of Artificial Intelligence I: Basic Techniques).

CSE 794: Foundations of Spoken Language Processing. Prerequisites for CSE 794 are CSE 625 (Introduction to Automata and Formal Languages) or Linguistics 684.01 (Introduction to Computational Linguistics I); CSE 630 (Survey of Artificial Intelligence I: Basic Techniques); CSE 730 (Survey of Artificial Intelligence II: Advanced Topics) or Statistics 428 (Introduction to Probability and Statistics II).

2.4 Relation to CSE and ABET objectives

1. Students will:
   1. Demonstrate proficiency in the areas of software design and development, algorithms, operating systems, programming languages, information systems, and computer architecture.
   2. Demonstrate proficiency in relevant aspects of mathematics, including discrete mathematics and probability, as well as electrical circuits and devices.
   3. Successfully apply these principles and practices to a variety of problems.
2. Students will:
   1. Demonstrate an understanding of differential and integral calculus, and of statistics.
   2. Demonstrate an understanding of the basic principles of physics and at least one other laboratory-based science.
   3. Demonstrate an understanding of the basic principles of at least one other engineering discipline in addition to computing and electrical engineering.
3. Students will:
   1. Demonstrate familiarity with basic concepts and contemporary issues in the social sciences and the humanities.
   2. Demonstrate an understanding of social, professional and ethical considerations related to engineering in general and to computing in particular.
4. Students will:
   1. Demonstrate an ability to work effectively in teams.
   2. Demonstrate an ability to communicate effectively.
5. Graduates will:
   1. Find suitable positions in industry and government that offer the prospect of challenging and rewarding careers in computing.
   2. Demonstrate an ability to acquire new knowledge in the computing discipline and to engage in life-long learning.
   3. [Graduates with an aptitude for, and interest in, graduate studies will] Apply to and be accepted for entry by strong graduate programs in computing.

2.3.1 CSE Outcomes

Criterion I

Criterion II

Criterion III

Criterion IV

Criterion V

2.3.2 ABET Criterion 3: Program Outcomes and Assessment

Engineering programs must demonstrate their graduates have:

   a.  an ability to apply knowledge of mathematics, science, and engineering
   b.  an ability to design and conduct experiments, as well as analyze and interpret data
   c.  an ability to design a system, component, or process to meet desired needs
   d.  ability to function on multi-disciplinary teams
   e  an ability to identify, formulate, and solve engineering problems
   f.  an understanding of professional and ethical responsibility
   g.  an ability to communicate effectively
   h.  the broad education necessary to understand the impact of engineering solutions in a global and societal context
   i.  a recognition of the need for, and an ability to engage in life-long learning
   j.  a knowledge of contemporary issues
   k.  an ability to use techniques, skills, and modern engineering tools necessary for engineering practice.

The AI courses contribute strongly to a number of ABET objectives. The AI courses are heavily oriented towards interdisciplinary education involving other engineering and scientific disciplines, problem solving through a variety of labs, and theoretical foundations.
 

2.5 Feedback

2.6 Changes since last report

CSE 630: Survey of Artificial Intelligence I: Basic Topics now incorporates a unit on introductory machine learning topics in order that students receive a broader introduction to AI, particularly since much of the field is moving towards machine learning techniques for complicated tasks. We no longer teach semantic networks as a topic.

CSE 634: Computer Vision for Human-Computer Interaction was developed to introduce the fundamental algorithms, concepts, and applications of computer vision to the AI curriculum. There previously was no computer vision course offered in the CSE department, though this area is a mainstream topic for AI and is taught at most major universities. Together with the ECE Department, we have successfully created a series of complementary courses related to this area.

CSE 730: Survey of Artificial Intelligence II: Advanced Topics now incorporates a final project and has been reorganized to highlight the research in the department through a combination of lectures by the primary lecturer and talks by guest lecturers illustrating approaches to research in AI. The audience of the course is about half undergraduates and half graduate students; the changes were primarily targeted for new AI graduate students (since this course is taught in the autumn quarter). However, several undergraduates commented on the course evaluations that they enjoyed having research topics integrated into the curriculum, as most coursework does not offer insights into the research done at OSU. In the AU03 offering, we noticed that many students had trouble formulating an appropriate research paper. Subsequently, in AU04 we added a small unit on how to write an AI research paper, using models from the guest lectures. This led to marked improvement in the final papers for the class.

CSE 732: Computational Linguistics was revived in SP04 with significant changes. The textbook has been changed to Jurafsky & Martin: Speech and Language Processing, which discusses a variety of techniques and has good coverage of the wide-ranging tasks that the field encompasses currently. The course content is targeted at advanced undergraduates and beginning graduate students in the artificial intelligence (CSE) or computational linguistics (LING) major. The course has been modified to include programming assignments in which the students build several simple components for a spoken dialog system. Based on current needs and our desire to distinguish this course from other offerings at OSU, this course has been refocused to concentrate on how knowledge representation and reasoning come into play during language processing, to expose the students to the special processing needs of interactive spoken language, and to provide a basic introduction to techniques for generation within interactive systems.

CSE 779: Introduction to Neural Networks has had a name change (from Introduction to Artificial Neural Network Methods) to reflect the more common usage of the term "neural networks". Coupled with the adoption of a new textbook, there is also more emphasis on statistical learning. This course has been cross-listed with ECE in the past, but there are now no faculty on the ECE side who are teaching this course, so this course is now being solely taught by CSE faculty.

CSE 794L: Foundations of Spoken Language Processing was piloted in Winter 2005; this course was developed to allow graduate and undergraduate students to interact in a team-based environment while learning about issues in automatic speech recognition and text-to-speech processing. The course attracted an interdisciplinary collection of students from CSE, ECE, and Linguistics.

2.7 Continuing concerns and future changes

For CSE 630: Survey of Artificial Intelligence I: Basic Topics, the faculty have been discussing the differential in backgrounds of students that come to CSE 630; both undergraduate and graduate students come from a number of different departments with varying expertise. This is in part because CSE 630 serves different purposes: as a technical elective in the CSE undergrad program, as a requirement in the Cognitive Science minor program, and as an intro for non-AI graduate students. Possible solutions include offering an honors section, or a gentler introduction (possibly listed as CSE 530), but these solutions require further discussion. We also, in preparing for this report, conducted an evaluation of how we felt the students were performing against the published outcomes for the course, which identified some outcomes that were outdated because of topic shifts in the course over time. Similarly, the course description included items no longer taught in the class. Thus, we are currently rewriting the outcomes and course description to better reflect the course content.

CSE 735: Methods in Pattern Recognition and CSE 776: Hardware/Software Interface Design Project have not been taught in recent history, but the AI faculty maintain interest in offering these courses. We expect that CSE 735 will be reactivated in the next few years with the hiring of Mikhail Belkin. We hope that CSE 776 can be reactivated if we are able to hire new faculty in related areas.

CSE 739: Knowledge-Based Systems in Engineering has been cross-listed with Chemical Engineering and Civil Engineering, and has most recently been taught by Adeli in Civil Engineering. Adeli recently approached the faculty to see about changing content of the course, which is still under discussion.

We hope to turn CSE 794L: Foundations of Spoken Language Processing into a permanent course if the subsequent pilots are successful.

3. Conclusions

People involved in preparing report: Jim Davis and Eric Fosler-Lussier, with comments from Donna Byron and Leon Wang.

Date of report: May 2005