home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!mips!sdd.hp.com!zaphod.mps.ohio-state.edu!rpi!uwm.edu!lll-winken!tazdevil!henrik
- From: henrik@mpci.llnl.gov (Henrik Klagges)
- Newsgroups: comp.ai.neural-nets
- Subject: ALN's FAIL ON 1-BIT ERRORS ! DISASTER IMMINENT ?
- Message-ID: <?.714341883@tazdevil>
- Date: 20 Aug 92 20:18:03 GMT
- Sender: usenet@lll-winken.LLNL.GOV
- Lines: 25
- Nntp-Posting-Host: tazdevil.llnl.gov
-
- [Just to retaliate a little bit against Bill's nn-bashing 8-)]
-
- The Neuronal Enquirer, 8/20/2002
-
- ALN's FAIL TERMINALLY ON SINGLE BIT ERRORS ! SCIENTISTS PARALYZED !
- According to informed sources, ALN's have a much hyped feature,
- lazy evaluation that allows the switchoff of subtrees of any size
- by a single logic gate ! Pathological information or single bit
- failures could easily disable the major parts of the ALN, thus
- loosing all the recognition capabilities of these subtrees.
-
- Numerous rumours indicate major impacts have already happened. The
- Pentagon's B2.5 - super stealth bomber (it is so stealthy that even
- at a price tag of $2e^9 it cannot be found in the federal budget !)
- is grounded ! Many more marriage rings than usual vanished in kit-
- chen sinks. U.S. trade negotiators were quoted 'We'll be real tough
- with them Japanese this time to bring our domestic semiconductor
- marketshare back over 7.5%'. The U.S. budget deficit for the first
- time exceeded the G.N.P., leading to the resignation of the complete
- board of Harvard Business school ...
-
- Cheers, Henrik
- IBM Research Division physics group Munich
- massively parallel group at Lawrence Livermore
-
-