"Genetic Programming : Artificial Nervous Systems, Artificial Embryos and Embryological Electronics", Hugo de Garis, in "Parallel Problem Solving from Nature", Lecture Notes in Computer Science 496, Springer Verlag, 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the evolution were usually taken. In the first approach, both operons were switchable during a run (i.e. the second operon was able to switch on later in the run) In the second approach, so called shaping techniques were used. The concept of shaping has been defined in earlier publications [de GARIS 1991a,b, 1992] Shaping is simply splitting up an evolutionary process into intermediate phases, with intermediate targets. The results of a previous phase (i.e. an evolved chromosome) are fed as starting conditions (i.e. initial chromosomes) for the next phase. This is done to reduce the size of ....

"Genetic Programming : Artificial Nervous Systems, Artificial Embryos and Embryological Electronics", Hugo de Garis, in "Parallel Problem Solving from Nature", Lecture Notes in Computer Science 496, Springer Verlag, 1991.

.... paper, will inspire later papers to be written on the embryological growth of (neuro) electronic circuits, thus creating a new speciality called Embryonics (Embryological Electronics) A device which GPs (i.e. evolves) a complex (neuro) electronic circuit has been called a Darwin Machine [de GARIS 1991a] by the author. It is envisioned that such machines may play a role in building artificial nervous systems for future artificial creatures (or biots, i.e. biological robots) The artificial embryos presented in this paper are two dimensional shapes formed by a colony of cells in ....

....ideas from the extensive work already done on Production and Classifier Systems into differentiable chromosomes. In order to give a more concrete example of what coded instructions might be like, a brief introduction to some earlier work of the author on Artificial Embryology will now be given [de GARIS 1991a, 1992b] Basically, the main idea of this earlier work was to evolve reproduction rules for cellular automata (CA) such that the final shape of a colony of cells attained, as closely as possible, some desired shape, such as a square, triangle, ellipsoid, tadpole, etc. The state of a cell in one ....

[Article contains additional citation context not shown here]

"Genetic Programming : Artificial Nervous Systems, Artificial Embryos and Embryological Electronics", Hugo de Garis, in "Parallel Problem Solving from Nature", Lecture Notes in Computer Science 496, Springer Verlag, 1991.

....low fitness values. The CA evolutionary techniques introduced in this paper need not be restricted to replicators. The author has already used Genetic Programming (GP) techniques to evolve artificial embryos using colonies of CA cells to specify shapes i.e. contiguous, occupied, 2 state cells [de Garis 1991, 1992, 1993] Applying GP methods to CAs may become useful when nanoelectronics starts producing quantum dot arrays [Kirk Reed 1992] where the scale of these next generation electronic devices is so small that quantum tunneling becomes their essential operating principle. Component ....

.... of these neurite network growing CA Machines, plus some microprocessors to both measure the fitnesses of the neurite networks (at performing some task) and to control the GA, we may build what the author calls a Darwin Machine (i.e. a specialised hardware device for Genetic Programming [de Garis 1991, 1993] One of the longer term aims of the new Evolutionary Systems Department at ATR, is to build artificial brains using perhaps neurite network modules grown on Darwin Machines. Hence the name Brain Builder Group . This group also includes a molecular programmer (i.e. someone who software ....

"Genetic Programming : Artificial Nervous Systems, Artificial Embryos and Embryological Electronics", Hugo de Garis, in "Parallel Problem Solving from Nature", Lecture Notes in Computer Science 496, Springer Verlag, 1991.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC