F. Gruau, "Genetic Micro Programming of Neural Networks," In Advances in Genetic Programming, K. Kinnear, Ed., Cambridge, MA:MIT Press, pp. 495-518, 1994.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....problems (see [5] for an extensive bibliography) However, only a very small number of results have been reported where GP, appropriately modified, has gone beyond the production of sequential tree like programs. For example, using cellular encoding GP has been used to develop neural nets [2] and electronic analogue circuits [4] while using interpreters implementing parallel virtual machines it has been used to develop special kinds of parallel programs [1, 7] This paper describes Parallel Distributed Genetic Programming (PDGP) a new form of GP which is specialised in the ....

F. Gruau. Genetic micro programming of neural networks. In K. E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24, pages 495--518. MIT Press, 1994.

....image analysis, etc. 11, 12, 9, 10, 1, 18] When appropriate terminals, functions and or interpreters are defined, stan dard GP can go beyond the production of sequential tree like programs. For example using cellular encoding GP can be used to develop (grow) structures, like neural nets [6, 7] or electronic circuits [15, 13] which can be thought of as performing some form of parallel analogue computation. Also, in conjunction with an interpreter implementing a parallel virtual machine, GP can be used to translate sequential programs into parallel ones [24] or to develop some kinds of ....

Frederic Gruau. Genetic micro programming of neural networks. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24, pages 495-518. MIT Press, 1994.

....image analysis, etc. Koza, 1992, Koza, 1994, K. E. Kinnear, Jr. 1994] When appropriate terminals, functions and interpreters are defined, standard GP can go beyond the production of sequential programs. For example using cellular encoding GP can be used to develop structures, like neural nets [Gruau, 1994] or electronic circuits [Koza et al. 1996] which perform some form of parallel analogue computation. Also, in conjunction with an interpreter implementing a parallel virtual machine, GP can be used to translate sequential programs into parallel ones [Walsh and Ryan, 1996] or to develop some ....

Gruau, F. Genetic micro programming of neural networks. In Kinnear, Jr., K. E., editor, Advances in Genetic Programming, chapter 24, p. 495--518. MIT Press.

....of an evolutionary algorithm is the genotype (the genome) and the executed solution is the phenotype (the living creature) this means that usage of a solution in nature denotes the life of a creature. But ontogenetic (and embryonic) Examples for these approaches are [Banzhaf, 1994; Gruau, 1994; Keller and Banzhaf, 1996; Koza et al. 1996; Eggenberger, 1996; Astor and Adami, 1998; Ferreira, 2001] development in nature happens within the lifespan of a create and not before the beginning of its being . There is one work that recognizes this di#erence and introduces many exciting ....

Frederic Gruau. Genetic micro programming of neural networks. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, pages 495--518. MIT Press, 1994.

....stimulus is more ordered, more nearly periodic during perception, than at rest. Cellular encoding systems. An alternative approach to evolving artificial neural networks which allows more complex circuitry with feedback loops was recently developed by Gruau. Cellular encoding [Gruau 1992a] Gruau 1994b] is a graph rewriting grammar expressed with a tree based representation. It is somewhat similar to the Genetic Programming Paradigm of Koza s school, see for example [Koza 1992] in that genotypes are recombined by swapping randomly chosen The evolution of artificial neurology 3 Figure 3 The ....

F. C. Gruau. Genetic micro-programming of neural networks. In Kinnear, editor, Advances in genetic programming, pp 495-518. The MIT Press, 1994.

....image analysis, etc. Koza, 1992, Koza, 1994, K. E. Kinnear, Jr. 1994] When appropriate terminals, functions and interpreters are defined, standard GP can go beyond the production of sequential programs. For example using cellular encoding GP can be used to develop structures, like neural nets [Gruau, 1994] or electronic circuits [Koza et al. 1996] which perform some form of parallel analogue computation. Also, in conjunction with an interpreter implementing a parallel virtual machine, GP can be used to translate sequential programs into parallel ones [Walsh and Ryan, 1996] or to develop some ....

Gruau, F. Genetic micro programming of neural networks. In Kinnear, Jr., K. E., editor, Advances in Genetic Programming, chapter 24, p. 495--518. MIT Press.

....image analysis, etc. 11, 12, 9, 10, 1, 18] When appropriate terminals, functions and or interpreters are defined, standard GP can go beyond the production of sequential tree like programs. For example using cellular encoding GP can be used to develop (grow) structures, like neural nets [6, 7] or electronic circuits [15, 13] which can be thought of as performing some form of parallel analogue computation. Also, in conjunction with an interpreter implementing a parallel virtual machine, GP can be used to translate sequential programs into parallel ones [24] or to develop some kinds of ....

Frederic Gruau. Genetic micro programming of neural networks. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24, pages 495--518. MIT Press, 1994.

....problems (see [5] for an extensive bibliography) However, only a very small number of results have been reported where GP, appropriately modified, has gone beyond the production of sequential tree like programs. For example, using cellular encoding GP has been used to develop neural nets [2] and electronic analogue circuits [4] while using interpreters implementing parallel virtual machines it has been used to develop special kinds of parallel programs [1, 7] This paper describes Parallel Distributed Genetic Programming (PDGP) a new form of GP which is specialised in the ....

F. Gruau. Genetic micro programming of neural networks. In K. E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24, pages 495--518. MIT Press, 1994.

....evaluation step can take many forms, while in GP an individual is evaluated for fitness at least in part by executing the program and by assessing the quality of its outputs. GP techniques have proven valuable for the evolution of structures other than computer programs (e.g. neural networks [1] and analog electrical circuits [2] but the emphasis on individuals as literal computer programs is the most central defining feature of GP. 1 Computational Universality The computational power of the set of elements out of which programs may be constructed the function set and terminal ....

F. Gruau, "Genetic micro programming of neural networks," in Advances in Genetic Programming, K. E. Kinnear Jr., Ed., pp. 495--518. MIT Press, 1994.

....the genetic algorithm. Gene duplication is implicitly used in the messy genetic algorithm (Goldberg, Korb, and Deb 1989) Lindgren (1991) analyzed the prisoner s dilemma game using an evolutionary algorithm that employed an operation analogous to gene duplication applied to chromosome strings. Gruau (1994) used genetic programming to develop a clever and innovative technique to evolve the architecture of a neural network at the same time as the weights are being evolved. 4.2. Argument Duplication The operation of argument duplication duplicates one of the dummy arguments (format parameters) in ....

Gruau, Frederic. 1994. Genetic micro programming of neural networks. In Kinnear, Kenneth E. Jr. (editor). Advances in Genetic Programming. Cambridge, MA: The MIT Press. Pages 495--518.

....weight connection matrix. Nolfi and Parisi (1995) use a more biologically based representation of network connectivity. The simulation of the phenomenon of cell division has been studied in different works that use the recursive mapping of Lindenmayer grammars. For example, in Belew (1993) and Gruau (1994) the final topology of the network is determined by the units duplications which are controlled by the rewriting rules encoded in the genotype. This kind of models of development has the limit of not dealing with an important aspect of the developing systems, namely the high interaction among the ....

F. Gruau (1994). Genetic Micro Programming of Neural Networks. In K.E. Kinnear (ed.), Advances in Genetic Programming. Cambridge, MA: MIT Press, Bradford Books.

....growth (Nolfi Parisi, 1995) The simulation of the biological developmental phenomenon of cell division has been studied in different works that use the recursive mapping of Lindenmayer grammars to determine the sequence and mode of unit proliferation in a neural network. In Belew (1993) and Gruau (1994) the final topology of the network is determined by the units duplications which are controlled by the rewriting rules encoded in the genotype. The advantages of using recursive mapping for cell duplication is shown to be relevant, especially for the development of complex neural structure, ....

Gruau, F. (1994). Genetic Micro Programming of Neural Networks. In K.E. Kinnear (Ed.), Advances in Genetic Programming. Cambridge, MA: MIT Press, Bradford Books.

....Its success is generally attributed to its use of an evolutionary search technique based on the genetic algorithm (GA) 4] combined with a dynamic, tree structured representation of the programs. Much recent research has focused on improving the performance of GP by improving its representation [2, 5, 6, 7, 17, 18], and this has proven to be an effective means of enhancing the basic algorithm. In theory the same effect as a change in representation could be achieved by a change to the transformation rules (in GA, the genetic operators) However, some effects may be more easily and practically achieved by ....

Gruau, F. (1994) Genetic Micro Programming of Neural Networks. In Advances in Genetic Programming, MIT Press, pp 495-518.

....image compression, image analysis, etc. 7, 8, 5, 6, 1, 11] When appropriate terminals, functions and or interpreters are defined, standard GP can go beyond the production of sequential tree like programs. For example using cellular encoding GP can be used to develop structures, like neural nets [3, 4] or electronic circuits [10, 9] which can be thought of as performing some form of parallel computation. Also, in conjunction with an interpreter implementing a parallel virtual machine, GP can be used to develop special kinds of parallel programs [2, 14] or to translate sequential programs into ....

Frederic Gruau. Genetic micro programming of neural networks. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24. MIT Press, 1994.

....function of automata in their entirety. Dunay et al. 1994] describes a genetic programming encoding which successfully produces generalizing automata for several languages, including all of the Tomita regular languages. This paper presents a method for applying cellular encoding, described by Gruau [1994], to the evolution of DFAs. The genetic programming tree is interpreted as a coding for the growth of an automata from an initial singlestate zygote. Mimicking the process of cell division in nature, execution of a GP tree causes the progressive division and development of cells in the embryo to ....

....TL5 pairwise, an even sum of ab s and ba s aabbaaaba aabbaaab TL6 number of b s number of a s = 3n (multiple of 3) abbaabbbb bbabaaa TL7 a b a b aababbb abbaabba Genetic Programming 1996: Proceedings of the First Annual Conference 2. Cellular Encoding of Deterministic Finite Automata Gruau [1994] presents a genetic programming technique, called cellular encoding, that allows for the concurrent evolution of the architecture, weights, and thresholds of a neural network. Each tree in the population represents a cellular code that specifies the development of a neural network from a single ....

[Article contains additional citation context not shown here]

Gruau, Frederic. (1994). Genetic micro programming of neural networks. Advances in Genetic Programming.

....image analysis, etc. 10, 11, 8, 9, 1, 17] When appropriate terminals, functions and or interpreters are defined, standard GP can go beyond the production of sequential tree like programs. For example using cellular encoding GP can be used to develop (grow) structures, like neural nets [6, 7] or electronic circuits [13, 12] which can be thought of as performing some form of parallel analogue computation. Also, in conjunction with an interpreter implementing a parallel virtual machine, GP can be used to translate sequential programs into parallel ones [22] or to develop some kinds of ....

Frederic Gruau. Genetic micro programming of neural networks. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24. MIT Press, 1994.

....image analysis, etc. 16, 17, 14, 15, 1, 22] When appropriate terminals, functions and or interpreters are defined, standard GP can go beyond the production of sequential tree like programs. For example using cellular encoding GP can be used to develop (grow) structures, like neural nets [9, 10] or electronic circuits [19, 18] which can be thought x y max 3 y x Figure 1: Parse tree representation of the expression max(x y, 3 x y) of as performing some form of parallel analogue computation. Also, in conjunction with an interpreter implementing a parallel virtual ....

....a common characteristic their offspring inherit such a characteristic, b) when parents have different characteristic their offspring can inherit both such characteristics, c) every offspring is a valid solution, d) crossover is efficient. Indirect graph representations, like cellular encoding [9, 10, 19, 18] or edge encoding [20] do not suffer from this problem as the standard GP operators can be used on them. However, such representations require an additional genotype to phenotype decoding step before the interpretation of the graphs being optimised can start, as the search is not performed in the ....

Frederic Gruau. Genetic micro programming of neural networks. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24. MIT Press, 1994.

....for developing a complete neural network from a very simple embryonic neural network (consisting of a single neuron) Genetic programming is applied to populations of these network constructing program trees in order to evolve a neural network capable of solving the problem at hand. See also Gruau 1994. Each program tree is a composition of network constructing, neuron creating, and neuron adjusting functions and terminals. The program tree is the genotype and the neural network constructed in accordance with the tree s instructions is the phenotype. The fitness of an individual program tree in ....

Gruau, Frederic. 1994. Genetic micro programming of neural networks. In Kinnear, Kenneth E. Jr. (editor). Advances in Genetic Programming . Cambridge, MA: The MIT Press. Pages 495--518.

No context found.

F. Gruau. Genetic Micro Programming of Neural Networks. Advances in Genetic Programming, 1994c. ed. Kim Kinnear, Mit Press.

No context found.

F. Gruau, "Genetic Micro Programming of Neural Networks," In Advances in Genetic Programming, K. Kinnear, Ed., Cambridge, MA:MIT Press, pp. 495-518, 1994.

No context found.

Frederic Gruau. Genetic micro programming of neural networks. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24. MIT Press, 1994.

No context found.

Frederic Gruau. Genetic micro programming of neural networks. In Kenneth E. Kinnear, Jr., editor, Advances in Genetic Programming, chapter 24. MIT Press, 1994.

No context found.

Frederic Gruau. Genetic Micro Programming of Neural Networks. In Kim Kinnear, editor, Advances in Genetic Programming. MIT Press, 1994.

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