COMP9414: Artificial Intelligence

Lecture Notes

Prolog Programming, Knowledge Representation and Machine Learning (5 weeks)

• Introduction to Artificial Intelligence
• Prolog and iProlog Programmer's Manual (Prolog on pc.solaris machines)
• [non-examinable: Enough Logic to Follow How Prolog Works]
• Production Rules and Prolog Code for Forward Chaining
• Uncertain Reasoning
• Belief Revision
• Frames lecture notes and
  On-line documentation of frames implementation in iProlog
• Problem Solving and Search, and Example of Prolog trace for path algorithm
• Introduction to Machine Learning
• Learning, Measurement and Representation
• Prototypes
• Function approximation
• Decision Tree Induction - ID3
• Neural Networks (Backpropagation) (postscript) or rough html version of backprop and XOR setup for iProlog
• Reinforcement Learning
• Genetic Algorithms
• Propositional Learning Systems
• Relations and Background Knowledge
• References

• Dictionary of Prolog Concepts (28 K)
• Dictionary of AI Concepts (34 K)
• Dictionary of Machine Learning Concepts (68 K)

Natural Language Processing (4 weeks)

• Linguistics
• Grammars and Parsing
• Features and Augmented Grammars
• Semantics and Logical Form
• Linking Syntax and Semantics
• Knowledge Representation for Natural Language
• Local Discourse and Reference
• Ambiguity Resolution - Statistical Methods

• Dictionary of Natural Language Processing Concepts (144K)

Lectures notes by A/Prof. Arcot Sowmya on...

Computer Vision (4 weeks)

• Introduction to Computer Vision
• Digital Images and Properties
• Pixel Relationships
• Binary Vision
• Preprocessing and Edge Detection
• Segmentation and Line Finding
• Region-Based Segmentation and Representation (covered in same document as previous section).
• 3D Segmentation and Description
• Model-Based Vision (covered in same document as previous section).
• Images Files

Bill Wilson | billw at cse.unsw.edu.au | Home Page
Last modified: 16 September 2002
CRICOS Provider No. 00098G