home *** CD-ROM | disk | FTP | other *** search
- Organization: School of Computer Science, Carnegie Mellon, Pittsburgh, PA
- Path: sparky!uunet!europa.asd.contel.com!emory!sol.ctr.columbia.edu!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!news.sei.cmu.edu!fs7.ece.cmu.edu!crabapple.srv.cs.cmu.edu!andrew.cmu.edu!at15+
- Newsgroups: comp.compression
- Message-ID: <8f0SkVm00iV14_M5MZ@andrew.cmu.edu>
- Date: Thu, 12 Nov 1992 00:31:13 -0500
- From: Andrew Lewis Tepper <at15+@andrew.cmu.edu>
- Subject: How I won $10
- Lines: 18
-
- A couple years ago I bet a friend that I could write a program to read
- his mind. Specifically, the computer would write "I predict you will
- type 0", or "I predict you will type 1". We would cover the screen up
- with a post-it note to prevent cheating. To win $10, the program would
- have to score 55% or better.
-
- My algorithm looked at past patterns of 1,2,3,etc bit sequences, and
- made it's prediction based on that. The computer got 569 out of 1000,
- and I won. Now, he was really trying to be "random". If you don't try
- too hard, the computer will often get 70-75%. My question is, is there a
- compression scheme that uses this technique? In certain kinds of data,
- bits could often be represented by (nearly) _zero_ bits of compressed
- data! Perhaps just a number that said "go on guessing for 136 bits".
- What kind of results do such algorithms achieve, if they in fact do
- exist?
-
- Andy
-
-