Learning Networks: An Alternative to Data Processing
A five papers session at the Northcon 1991
conference
Published / Presented at: Northcon/91 Conference Record, Technical Program Session D4, Wednesday October 2, 1991, Oregon Convention Center, Portland, Oregon, USA. Organizer: Klaus Holtz, Omni Dimensional Networks.
Download Conference Top Pages – Adobe pdf
Paper 1: AN
INTRODUCTION TO AUTOSOPHY AND SELF-GROWING NETWORKS
Author: Klaus Holtz, Pres. Omni Dimensional Networks, 631 O’Farrell #710, San Francisco, CA 94109, Tel. 415 474-4860.
ABSTRACT: AUTOSOPHY (Auto = self, SOPHY = knowledge or wisdom) is a scientific discipline originally conceived to explain the functioning of self-assembling natural structures, such as crystals, trees or societies. The laws and algorithms discovered during this research are used in Electronic Autosophy (EA) to grow self-assembling data networks, like “Data Crystals” or “Data Trees”, in electronic memories without programming or conventional data processing beyond the original set up of the learning algorithms. The resulting Electronic Autosopher duplicates the learning processes in the brain at the mathematical level. “Learning” is a new process that requires new mathematics and a new information theory.
Paper 2:
SELF-GROWING NETWORK TREES SIMPLIFY KNOWLEDGE BASE DESIGN
Authors: Klaus Holtz, Eric Holtz, Omni Dimensional Networks, 631 O’Farrell #710, San Francisco, CA 94109, Tel. 415 474-4860.
ABSTRACT: Self-growing Network Trees store and retrieve text data more efficiently and without conventional programming. The growth of the data trees can be compared to the growing of information crystals in a computer memory. A system may be educated very much like a human child to answer questions in grammatical language. The knowledge base information will be stored in a highly compressed and universal format for automatic merging with other systems. Searching for any information is nearly instantaneous and independent of the amount of store records.
Paper 3: SELF-ASSEMBLING DATA TREES YIELD NEAR
OPTIMUM DATA COMPRESSION AND ENCRYPTION
Authors: Klaus Holtz, Alfred Lettieri, Omni Dimensional Networks, 631 O’Farrell #710, San Francisco, CA 94109, Tel. 415 474-4860.
ABSTRACT: String compression, such as the Ziv Lempel codes, are used to compress files and speed communications, but limited encoding speed makes them unsuitable for very high speed operations such as ISDN or LANs. Codebook encryption may yield near unbreakable codes but suffer the same speed limitations. Very high-speed text compression and encryption are now made possible by “Autosophy” data trees, which grow, from text input, like “Information Crystals” in a memory without programming. AUTOSOPHY (AUTO = self, SOPHY = knowledge or wisdom) is a new science dealing with self-assembling structures such as crystals, living trees, societies or self-growing information networks.
Paper 4: HIGH
SPEED VIDEO COMPRESSION FOR HDTV AND MULTIMEDIA IMAGE STORAGE
Author: Klaus Holtz, Omni Dimensional Networks, 631 O’Farrell #710, San Francisco, CA 94109, Tel. 415 474-4860.
ABSTRACT: Autosophy based television extracts the “True Information” from the images for storage or transmission, resulting in enormous data compression and optional encryption for security. True information in television images is not dependent on screen size, resolution or scanning rates but rather only on the “Novelty” and movement within the images. Image transmission or storage is with “Superpixel”, which may represent any size image fragment, from single pixel to entire screen images. True information is extracted from the television images by comparing the image, like a jigsaw puzzle, with previously learned images, a process that is practical only with Autosophy.
Paper 5: MACHINE
VISION WITH SELF-GROWING NETWORKS
Author: Klaus Holtz, Pres. Omni Dimensional Networks, 631 O’Farrell #710, San Francisco, CA 94109.
ABSTRACT: The emerging science of AUTOSOPHY (AUTO = self, SOPHY = knowledge or wisdom) explores self-assembling structures such as crystals, living trees or societies. It provides basic laws and algorithms for growing self-assembling data networks, like “Information Crystals”, in a memory without programming beyond the set up of the learning algorithms. Autosophy based vision is self-learning and similar to our own brain. Recognition speed is independent of the number of objects in the target library. Images are learned by exposing models or photographs to the camera in various orientations.