The following figure summarizes the courses and their prerequisites, as of the summer of 1997.
(CMSC 330) (CMSC 251 or CMSC 420)
\ /
\AND/
\ /
CMSC 421, Introduction to Artificial Intelligence
/ / | \ \ \
/ / | \ \ \
/ / | \ \ \
/ / | \ \ \
CMSC 720, CMSC 721, CMSC 722, CMSC 723, CMSC 726, CMSC 727,
Logic for Non-Monotonic AI Natural Machine Neural
Problem Solving Reasoning Planning Language Learning Modeling
CMSC 421, Introduction to Artificial Intelligence (3 credits). Prerequisites: CMSC 330 and either CMSC 251 or CMSC 420, or equivalent (or permission of instructor). Areas and issues in artificial intelligence, including search, inference, knowledge representation, learning, vision, natural language, expert systems, robotics. Implementation and application of AI programming languages such as Lisp and Prolog.
CMSC 720, Logic for Problem Solving (3 credits). Prerequisites: CMSC 421 or equivalent (or permission of instructor). Logic programming and its use in problem solving, natural language recognition and parsing, and robotics. The Prolog language. Meta-level and parallel logic programming. Expert systems. Term project in logic programming.
CMSC 721, Non-Monotonic Reasoning (3 credits). Prerequisites: CMSC 421 or equivalent (or permission of instructor). This course will survey several of the major standard formalisms for nonmonotonic reasoning, and also look at current research issues.
CMSC 722, Artificial Intelligence Planning (3 credits). Prerequisites: CMSC 421 or equivalent (or permission of instructor). A long-standing problem in AI is how to plan a set of actions to accomplish some desired goal. This course will cover the basic algorithms, important systems, and new directions in the field of AI planning systems.
CMSC 723, Natural Language Processing (3 credits). Prerequisites: CMSC 421 or equivalent (or permission of the instructor). Introductory course on applications of computational techniques to linguistics and natural language processing. Research cycle of corpus selection, pre-editing, keypunching, processing, post-editing, and evaluation. General purpose input, processing, and output routines. Special purpose programs for sentence parsing and generation, segmentation, idiom recognition, paraphrasing, and stylistic and discourse analysis. Programs for dictionary, thesaurus, and concordance compilation and editing. Systems for automatic abstracting, translation, and question answering.
CMSC 726, Machine Learning (3 credits). Prerequisites: CMSC 421 and an undergraduate course in elementary probability/statistics (or permission of the instructor). Reviews and analyzes both traditional symbol-processing methods and genetic algorithms as approaches to machine learning (neural network learning methods are covered in CMSC 727). Topics include induction of decision trees and rules, version spaces, candidate elimination algorithm, exemplar-based learning, evolution under natural selection of problem-solving algorithms, system assessment, and comparative studies.
CMSC 727, Neural Modeling (3 credits). Prerequisites: CMSC 421 or equivalent (or permission of instructor). Undergraduate calculus, linear algebra, and elementary probability theory are assumed. Fundamental methods of neural modeling. Surveys historical development and recent research results from both the computational and dynamical systems perspective. Logical neurons, perceptrons, linear adaptive networks, attractor neural networks, competitive activation methods, error back-propagation, self-organizing maps, and related topics. Applications in artificial intelligence, cognitive science, and neuroscience.