| TTL Databook of the Mind (2008) | |||||||||||||||
Abstract | |||||||||||||||
| The Problem: This is a long-term effort to develop components which can serve as a foundation from which complex “intelligent ” systems can be built, much the way that the TTL databook served as a foundation from which complex digital electronic systems could be built. Wewanttohaveasimilar library of modular components, each able to perform some well-specified aspect of “intelligent ” behavior. Motivation: Recent research suggests that the human brain may be a highly organized structure with many specialized components, rather than some sortofvastdisorganized “neural pudding. ” The human genetic code, however, is only about 1 gigabyte, too small to encode the full structure of the brain in detail. One way of reconciling the structure of the brain with the poverty of instructions available for constructing it is to imagine the brain as composed of a large number of small components, where each component has a well-specified design drawn from arelatively small library of component types. In building a TTL databook for the mind, we hope to attack the problem of intelligence on a median level of complexity: high-level in the sense that we are not attempting to model neural hardware, but are concerned with pieces of functionality, low-level in thesense that we are concerned with building simple mechanisms capturing only small fragments of “intelligence”. From the first simple components, we hope to build more complex components, then more complex components | |||||||||||||||
Publication details | |||||||||||||||
| |||||||||||||||