Description
Real-time information processing structures perform critical duties in natu- ral systems ranging across gene expression, cellular metabolism and repair, the immune response, neural process, small-group `cockpit’ actions, and, for humans and other colonial living things, the operation of elaborate `insti- tutions’ of various forms, dynamics, and purposes. Draconian evolutionary selection has determined the form, function, and stabilizing mechanisms of such structures over the billion years of our interacting biological, social, and, more recently, cultural histories. In spite of this, natural real-time cognitive systems fail. We age, become ill, and die, and our socioeconomic institutions collapse, in large measure consequent on the failure of critical regulatory processes (Wallace 2015a, b).
Now, we are entering a new evolutionary dynamic { sometimes called `anthropocene’ { dominated by articial cognitive structures at varying scales and levels of organization. Evolutionary selection pressures are poised to act on the 7 billion humans enmeshed within those structures. The purpose of this book is to develop a set of information theoretic statistical tools for analyzing the instabilities of real-time cognitive systems at those varying scales and levels of organization, with special focus on high level machine function.
In particular, technological momentum, driven by relentless economic interests, will see massively parallel computers widely deployed to control a great swath of critical real-time phenomena ranging across transportation, power, nancial, and communication networks, individual vehicles, chem- ical production facilities, reneries, nuclear reactors, and the like. It is increasingly understood that stabilizing cognitive systems and devices is as inherently intractable as programming them { something we do not really know how to do well. This book formalizes that understanding for cognitive