Resources : Artificial Intelligence : Featured Resources
Bit Code Examples:
Bit Code is an attempt to make a Turning Complete language with minimal commands; in fact, only two: one to increment a number pointed to by a data address, and one to load a new value into the data address. For added cleverness, the data pointed to by the data address is used as the pointer to the next command, so that incrementing this number also moves on to the next command. I have a C program for generating and running random strings of bits and a GIF and FIG graphics that explain how the syntax works.
Java/Scheme Interface, for creating a dual-language system. Logically, Java keeps track of objects and data, and Scheme tracks operations and decisions (possibly with a codelet system). Here is the Java Code and the Scheme Code.
Scheme Word
Histogram: finds words that are sufficiently "unnotably" in
several files (common words, with more than one character, but only
very common words if they're long).
Bit Code Examples:
Bit Code is an attempt to make a Turning Complete language with minimal commands; in fact, only two: one to increment a number pointed to by a data address, and one to load a new value into the data address. For added cleverness, the data pointed to by the data address is used as the pointer to the next command, so that incrementing this number also moves on to the next command. I have a C program for generating and running random strings of bits and a GIF and FIG graphics that explain how the syntax works.
Artificial Intelligence Workshop Description: in PDF, DVI, and LaTeX.
Sinusoid-Based prediction "AI". As it
gathers more information, it tries to randomly modify its predictive
equations, which is a Fourier series to better predict the input. The
C file needs an input file
input.txt to predict. The FourierAt function looks
wrong...
Image
Recognition System (still buggy) which determines "example-ness"
by clustering based on responses to filters. Includes Pattern
(responses to filters used for clustering),
Cluster (combination of Patterns), ClusTree (clustering
using nearest-neighbor trees), Image (ppm file reader),
GausImage (image on multiple scales for multidimensional
pattern recognition), Histogram (a collection of clusters for
statistical analysis; contains multiple clustering algorithms),
Model (multiple histograms describe a class of images), and
System (a combination of models for classifying an image), and
Cluster data file I/O.
Resources : Artificial Intelligence : Links to Other Sites
arXiv e-Print Archive Repository of repositories of papers including abstracts and full PDFs.
If you have any questions or comments that you want to email, feel
free to contact me, at jrising@ mit.edu or with the Comments and
Questions Form. If you find any dead links, please
tell me (I will fix them)!