pizza / genx
- Source
- Commits
- Network (0)
- Issues (0)
- Downloads (1)
- Wiki (1)
- Graphs
-
Branch:
3-params
genx /
| name | age | message | |
|---|---|---|---|
| |
ABOUT.txt | ||
| |
INSTALL.txt | ||
| |
LICENSE.txt | Fri Feb 27 01:03:45 -0800 2009 | |
| |
README | ||
| |
TODO.txt | ||
| |
doc/ | ||
| |
misc/ | ||
| |
src/ |
README
# ex: set ff=dos ts=2 et: # $Id$ o=O 0===0 0===0 README O=o ---------------------------------------------------------------------- O o=O For the impatient: 0===0 git clone git://github.com/pizza/genx.git ~/genx && cd ~/genx/src && make && ./genx 0===0 sudo renice +19 $(pgrep genx) O=o O o=O genx generates functions in x86 machine language via a genetic 0===0 algorithm. 0===0 O=o It produces genes consisting of random machine language. O These genes are fed input and their output is compared to a target; o=O for each generation of genes those most closely matching the target 0===0 output are kept and used to populate the next generation. 0===0 O=o Eventually the gene converges upon a function that maps accurately O between the target input and output. o=O 0===0 0===0 Why is this interesting? O=o ---------------------------------------------------------------------- O o=O Discover Unknown Functions/Relationships 0===0 You can feed the program data whose statistical relationship is 0===0 unknown and it can search for relationships in an automated O=o fashion. O o=O Automated Optimization 0===0 Find shorter and more efficient methods of known functions. 0===0 O=o It's Fast O Because the functions are expressed in native x86 they can be o=O executed by the CPU directly, allowing even a very modest desktop 0===0 computer to try many millions of combinations per minute. 0===0 O=o O Further Information o=O ---------------------------------------------------------------------- 0===0 0===0 INSTALL.txt <- Figure out how to use it O=o TODO.txt <- Help out O LICENSE.txt <- Concerning distribution o=O 0===0 0===0 O=o O o=O 0===0 0===0

