Heuristic Computing by Dr. Yoshiyasu Takefuji, Professor
What you can learn: In this class, you can learn automated reasoning (otter) and neural computing by solving amusing puzzles and brain teasers. You will be able to build a highly intelligent system based on the given schemes. Heuristic functions including greedy algorithm and A* algorithm will be briefly addressed. In the first class, solving a variety of puzzles will be demonstrated. Open source languages including python, C, otter, and prolog are used in the class. You can learn state of the art on the artificial learning.
What you have to do: You have to submit your homework every week. A small pop-quiz will be given every week.
In the last week you have to give your presentation and report on your final mini-project.
Requirements: You have to have a notebook of Windows, Linux or Mac.
Open source software used in the class:
Python2.6,
neurolab,
pybrain,
ECLiPSe,
SWI-Prolog,
otter,
gcc
Puzzles used in the class:
8-queen problem, knight's tour,
four coloring map,
fifteen_puzzle,
knights&knaves&spies puzzle,
Missionaries and Cannibals Puzzle,
2-inverter puzzle,
Twelve Billiard Balls Puzzle, and other intractable puzzles.
References:
Neural Network Parallel Computing
Neural Network Introduction in JAPANESE
backpropagation
automated reasoning
otter
Other references(PDFs): Ajax for Dummies Android CookBook Data Structures and Algorithm using Python Discrete Optimization Heuristics How to solve it modern heuristics Making games Prolog Lab Book09 Python for Dummies Python algorithms Algorithmic Problem Solving with Python Andoid Developer's Cookbook