Annual Conference CybCon 2009
Abstracts and supporting media received
"Complexity from simplicity: the nature of light and matter"
Dr. John Williamson writes:
It is well known that matter can be created from pure energy. In particle physics experiments this has become routine enough that it is now largely taken for granted. Despite this just quite how energy, a simple form, becomes enfolded into the complexity of matter and how then these complex forms interact amongst each other is still less than wholly clear.
A new basis for physics is proposed which provides a possible route by which certain aspects of this may be understood. Starting from the simple and continuous, topological soliton solutions are proposed as quantised matter. A new class of solutions to electromagnetism corresponding more closely than the usual forms to quantised photons, will also be explored.
Foils from (CybCon08) presentation in .pdf document.
Download the foundation paper "Is the electron a photon with a toroidal topology?" Williamson and van der Mark, Annales de la Fondation Louis de Broglie, Volume 22, no.2, 133 (1997).
Draft paper on new theory "electremdense2008v4.pdf" for Kybernetes.
"In the beginning was the Turtle"
Artemis Papert writes:
Around 40 years ago Seymour Papert invented the now famous Logo Turtle. That point in the 60s was the middle of a history rather that the beginning. Twenty years earlier was the wonderful Grey Walter Turtle. In the 70's the Logo screen Turtle appeared, jumping from the floor to the computer screen. Most recently the Turtle found its way into TurtleArt.
This talk will describe this history, demonstrate the art, math, and programming of TurtleArt, and touch on some of the ways the Turtle has influenced our lives and learning.
" 'Observation operators' revisited"
Professor Robert Vallée writes:
We consider a macroscopic dynamical system S able to perceive the evolution of its environment and itself, and so the evolution of the state of the “universe”, described by function x. The evolution of its perceptions is given by function y and we have y = O(x). We call O the “observation operator” (Vallée 1951). It acts only on the past and present of x. When x and y belong to linear spaces X and Y = O(X), operators O may be linear (Volterra composition) with two extreme cases y(t) = A(t) and y(t) = ∫Z(t-s) x(s) ds (convolution with Z(t-s) null for s>t). When A(t) is a mere scalar we may have a factor of attention or a time window. If Z is a scalar we have a frequency filter, If Z(t-s) = δ(t-s) we have a time lag operator.
The concept of observation operator gives a description of “epistemological indiscernibility” which happens when O has no inverse, in such a case we may have O(x') = O(x) with x'≠ x. The “resolving power” of the observation operator O may be defined by the set of all the equivalence classes x* of O (all the elements of x* give the same perception y).
When we have a mapping f of the set A on set B, and a structure on B, there is a classical process to obtain on A an induced structure involving the reciprocal image mapping f-1 which exists even when the inverse mapping does not. Léon Motchane (1958) gave an interpretation of this process in the field of measurement. We integrated it in the framework of our dynamical formalism of observation operators. For example if Y = O(X) is structured by a binary relation R, a corresponding binary relation T is induced on X, defined by xTx' meaning that O(x) R O(x'). So x and x' are subjectively perceived by S as related by T when their images are related by R. It is very natural because the only relations S can know are those in Y. We call this situation “epistemological transfer of structure” (Vallée 1974, rediscovered by Robert Rosen 1978). Analogous remarks may be made about the transfer of topology or measure.
If we suppose that system S is also able to decide, we must have a process of decision described by a decision operator D (in a first approach acting automatically). So if x generates y = O(x), y will generates z = D(y) = P(x), P being what we call “pragmatic operator” (Vallée 1974). The properties of P are identical to those of O (“pragmatic indicernibility”, “pragmatic inverse transfer”). Function z gives the evolution of decisions taken by system S, it is, as well as y, a description of the way S perceives the universe: description by decision taken instead of by direct perceptions felt. We call this way of seeing epistemology “epistemo-praxiology” (Vallée 1987).
"Strong AI, Continuity and Evolution"
Dr Alex Andrew writes:
The "strong-AI" position has been epitomized in a claim that there is no essential difference, except complexity, between a room thermostat "knowing" that a room is warm or cold, and a person making the same discrimination. The position will be defended, though with admission of vagueness arising from a qualifier such as “essential”.
Nevertheless it is widely acknowledged (even by Marvin Minsky) that results from the AI effort up to now are disappointing, and reasons will be discussed. One is of course simply the brain's extreme complexity and versatility, but particular attention will be paid to the part played by continuous processing in mentation, as treated in the speaker's recent book (A Missing Link in Cybernetics: Logic and Continuity, Springer 2009). Relevance of the principles to biological evolution will be discussed.
"Cybernetics as Science of Decision-making"
Dr Helmut Nechansky writes:
Control results from the interaction of two systems, a controller and a controlled system. And the main internal activity of a controller is to make a decision for a certain action. If such a decision actually leads to control, is a question of the external interaction of the selected action of the controller with the controlled system. Therefore cybernetics has to deal, too, with the question, what kind of decisions a controller has to make, so that control can take place.
The detailed argumentation to illustrate that starts with an analysis of the structure of a feedback system and the decisions this structure enables. From that the following main hypothesis are derived:
Then it is shown in a few examples how feedback systems can be enlarged. E.g. a second sensor can be added with a few changes to the structure. Based on that further additions can enable recognition of changes in space or in time. These examples illustrate both hypotheses, how more complex controller structures enable certain complex decisions.
As an application of these steps the controller structures necessary for Miller's living systems and for Beer's viable systems are analyzed and presented in one comparable scheme. That enables to show the relationship between Miller’s and Beer’s theory, which had not been explained previously.
Finally an outlook is given on forthcoming work, how this approach can be used to surface the cybernetic necessities of increasingly complex structures, like adaptive systems, learning systems, etc. and the decisions made by such systems.
So the paper seeks to forward an understanding of cybernetics as a science of structures that enable certain decisions. It is suggested, that making explicit this aspect of decision-making gives cybernetics an additional relevance for social sciences, like psychology, organization theory and sociology.
Papers supporting Cybernetics of Decisions
"A Tour of European Historical Robots and Cybernetic Art"
Martin Smith writes:
David Buckley recently returned from a five week European tour of museums, institutes and collections of robots, automata and cybernetic art works. His tour took in the Burden Neurological Institute where W. Grey Walter developed the first turtles. He photographed and video recorded many items in places including Berlin and Leipzig. His illustrated talk will describe and place in context what he saw and learned on his "pilgrimage".
"Science and Design of Systems"
Martin Smith writes:
Science describes the world in terms of phenomena classified according to shared properties, which are used to create mathematical relations that translate notions into refutable relationships by exposing them to the test of experience. There is another view of the world, the view of related objects, the 'systemic view'. The aim of this research is to show how to convert the systemic view into systems science using the scientific method to model objects (natural, technical, living, human etc) and their activities. Having identified the fundamental notions, by means of energy or information flow, they are organised into inferential structures using mathematics of 'ordered pairs' and predicate logic statements. These structures enable energy and information to propagate, to produce an outcome.
Jasia Reichardt writes:
The Frankenstein Monster: 200 years old and still relevant. Whether made out of meat, plastic, electronic components or on the internet, the generation of new beings gives us more problems than we can solve.
The Archbishop of Canterbury had an idea...
About our speakers
King's College street map by Multimap
Car travellers can park nearby in Surrey Street, Temple Place or Arundel Street at parking meters which are free after 6.30 pm. The congestion charge does not apply after 6.30 pm. Before 6.30 the charge is £8 (if paid before 10 pm). It can be paid by telephone on 0845 900 1234 or on the Congestion Charge website.
For Underground train travellers the nearest stations are Temple (500 metres, District and Circle lines), Covent Garden (800 metres, Piccadilly line), Embankment and Charing Cross (800 metres, Northern and Bakerloo lines).
Bus travellers may catch the 6, 7, 13, 23, 76, 521 and RV1.