J. R. Koza, F. H. Bennet III, D. Andre, M. A. Keane, and F. Dunlap, "Automated synthesis of analog electrical circuits by means of genetic programming," IEEE Transactions on Evolutionary Computation, vol. 1, no. 2, pp. 109--128, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....guide the evolution of candidate designs toward a suitable result in a reasonable time. There have been a number of research efforts aimed at exploring the combination of genetic programming with physical modeling to find good engineering designs. Perhaps most notable is the work of Koza et al. [2 4]. He presents a single uniform approach using genetic programming for the automatic synthesis of both the topology and sizing of a suite of various prototypical analog circuits, including low pass filters, operational amplifiers and controllers. This system has already shown itself to be extremely ....

Koza,, J.R., Bennett, F.H., Andre, D., Keane, M.A., and Dunlap, F., 1997, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Trans. Evol. Computation, 1(2), pp.109-128.

....guide the evolution of candidate designs toward a suitable result in a reasonable time. There have been a number of research efforts aimed at exploring the combination of genetic programming with physical modeling to find good engineering designs. Perhaps most notable is the work of Koza et al. [3, 4, 5]. He presents a single uniform approach using genetic programming for the automatic synthesis of both the topology and sizing of a suite of various prototypical analog circuits, including low pass filters, operational amplifiers, and controllers. This approach appears to be very promising, having ....

Koza,, J.R., Bennett, F.H., Andre, D., Keane, M.A., and Dunlap, F., 1997, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Trans. Evol. Computation, 1(2), pp.109-128.

....computation have appeared over the last few years. These include the use of genetic algorithms (GAs) 5] to select filter component sizes [6] to select filter topologies [3] and to design operational amplifiers using a small set of topologies [10] The research of Koza and collaborators [8] on analog circuit synthesis by means of genetic programming (GP) is likely the most successful approach to date. Unlike previous systems, the component values, number of components, and the circuit topologies are evolved. The genetic programming system begins with minimal knowledge of analog ....

....knowledge of analog circuit design and creates circuits based on a novel circuitencoding technique. Various analog filter design problems have been solved using genetic programming (e.g. 9] and an overview of these techniques, including eight analog circuit synthesis problems, is found in [8]. A comparison of geneticbased techniques applied to filter design appears in [14] and work on evolving CMOS transistors for function approximation [12] has also recently appeared. The system we present here was motivated by the genetic programming system described above. Our investigation ....

J.R. Koza, F.H. Bennett, D. Andre, M.A. Keane, F. Dunlap, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Trans. on Evolutionary Computation, vol. 1, no. 2, July, 1997, pp. 109--128.

....in inductive learning has long been recognized in the machine learning field and will not be repeated here. 3) Evolving Analogue Circuits: In comparison with digital circuits, analogue circuits are more difficult to design. Recent work on evolving analogue circuits using GA s [19] and GP [20] [24] shows an alternative to analogue circuit design using the evolutionary approach. One of the key issues in such evolutionary design is to find a suitable chromosome representation of analogue circuits. This problem is quite similar to that in evolutionary artificial neural networks (EANN s) ....

....similar to that in evolutionary artificial neural networks (EANN s) 25] 27] in which a good chromosome representation of EANN s is also very important. In the GP approach to analogue circuit design, trees are used to construct circuits. These circuit constructing trees are evolved by GP [20] [24]. Each tree can contain connectionmodifying functions, component creating functions, and automatically defined functions. A number of circuits, such as a low pass brick wall filter, an asymmetric bandpass filter, and an amplifier using transistors, have been evolved successfully [20] 23] The ....

J. R. Koza, F. H. Bennett, III, D. Andre, M. A. Keane, and F. Dunlap, "Automated synthesis of analog electrical circuits by means of genetic programming," IEEE Trans. Evol. Comput., vol. 1, pp. 109--128, Feb. 1997.

....following the instructions it grows the fully developed topology. It can even include the development for the intemal parameters associated with the functional elements. Problems similar to this one that involved development of topology graphs from expression trees were suggested by Koza et al. [6, 8] that used a mapping into analog circuit topologies; and Gruau [9] that mapped trees into neural network families. 2.2. Suggested mapping: expression tree into topology graph The nodes in the expression trees are instructions that when executed will result in a fully developed topology graph. ....

J.R. Koza, F. H. Bennett, HI, D. Andre, M. A. Keane, and F. Dunlap, "Automated synthesis of analog electrical circuits by means of genetic programming," IEEE Transactions on Evolutionary Computation, vol. 1, pp. 109- 128, 1997.

....process. In this process, thousands of parameters are combined in a complex way to produce a desired sound result. Following this original work, a series of articles related to the production of sound synthesis using GAs to deal with the complexity of an automated synthesis process were created [7][8][2] In recent literature, GAs have also been applied to produce evolving trajectories of musical material [6] Biles [1] presented a genetic algorithm based model that mimics a student learning to improvise jazz solos under the guidance of a human mentor. In Horowitz s [5] development, an ....

J. R. Koza, F. H. Bennet III, D. Andre, M. A. Keane, and F. Dunap. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation, 1(2), 1997.

....evaluated and refined in accordance with the filter type and reference signal in real time. This places a requirement that fitness evaluation is faster than the data rate of the audio signals. A number of researchers have looked at the problem of filtering using evolutionary methods [FS98, ZPV99, Kea97] 4.4 Non constrained Evolution If no limits are set as to how evolution may use the underlying technology then new solutions may be found that are not achievable with traditional (constrained) design methods. Evolution may use characteristics and parameters of the underlying technology which ....

J.R. Koza et al. Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming. IEEE Transactions on Evolutionary Computation, 1(2) pp 109--128, July 1997.

....Most work in this area has focused on analogue filter designs, particularly non adaptive filters which may be classified as passive or active. Genetic Programming (GP) 4] has been used to evolve active filter designs through the choice of topologies, component types and their respective values [5]. Evaluation of the evolved circuits was conducted using SPICE simulations. Zebulum et.al. have also evolved filter designs but in their case passive filters were the focus and a genetic algorithm was used instead of a genetic program [13] Both of these works may be classified under extrinsic ....

J. Koza et al. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation, 1(2):109--128, July 1997.

....building blocks are more likely to exhibit the desired stable behaviour. One way of exploring the use of evolution for design of analogue systems is extrinsic evolution in which an evolutionary algorithm is used to suggest circuits and a software simulator to test these candidates, such as in [3]. In our work we are looking at evolution as a method for obtaining the set of instructions to apply to programmable analogue devices. These are devices which have been designed to process analogue signals, but the nature of the analogue operations that are performed on the signals are specified ....

. Koza, John R., Bennett III, Forrest H, Andre, David, Keane, Martin A, and Dunlap, Frank. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation. 1(2) (1997) 109-128.

.... [7] to select filter topologies [4] and to design operational amplifiers using a small set of topologies [10] Various analog filter design problems have been solved using genetic programming, and an overview of these techniques, including eight analog circuit synthesis problems, is found in [9]. The remainder of the paper is as follows. First we discuss the genetic algorithm and the parallelization technique implemented in our software. Then we present the genetic representation of analog circuits and describe the genetic algorithm that is used. The design tasks and the experimental ....

....the target output. These error values were summed across evaluation points to arrive at a fitness value. The target specifications for this experiment was: f p = 1000 Hz, f s = 2000 Hz, K p =0.01dB, K s = 63.50 dB. These specifications are similar to the fifth order elliptic filter described in [9]. In that work, the evolved LC circuit satisfies K p =0.3dB and K s =60dB. Another evolved low pass filter circuit [19] had the same stopband and passband frequencies, but less demanding attenuation specifications (K p =1.6dB and K s =24.8dB) The evolved circuit is shown in Fig. 6(a) and its ....

[Article contains additional citation context not shown here]

J.R. Koza, F.H. Bennett, D. Andre, M.A. Keane, F. Dunlap, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Trans. on Evolutionary Computation, vol. 1, no. 2, July, 1997, pp. 109--128.

.... filter topologies [3] and to design operational amplifiers using a small set of topologies [12] Various analog filter design problems have been solved using genetic programming (e.g. 11] and an overview of these techniques, including eight analog circuit synthesis problems, is found in [10]. In space applications, automated evolutionary design of analog circuitry could hold many benefits, especially for controller hardware. In problems where actuator outputs need to be rapidly modulated in response to sensor feedback, analog circuits have some clear advantages over digital ....

.... Filter The third low pass filter design task had specifications that were the most stringent: in addition to the passband and stopband being increased, the attenuation parameters were tightened (see Table II) These specifications are similar to the fifth order elliptic filter described in [10]. In that work, the evolved LC circuit satisfies K p =0.3dBandK s =60dB.Another evolved low pass filter circuit [23] had the same stopband and passband frequencies, but less demanding attenuation specifications (K p =1.6dBandK s =24.8dB) The evolved circuit is shown in Fig. 12 and its frequency ....

[Article contains additional citation context not shown here]

J.R. Koza, F.H. Bennett, D. Andre, M.A. Keane, F. Dunlap, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Trans. on Evolutionary Computation, vol. 1, no. 2, July, 1997, pp. 109--128.

....of a game As this seems next to impossible, our solution is to replace the suite of hand coded behaviors with a complex fitness function that provides reward for good play even when no goals are scored. As Koza, Andre, Bennett, and Keane have noted in their research on evolving analog circuits [Koza et al. 1997] , it is significantly easier to write specifications for complex behavior than to write the programs to achieve the behavior. As part of the specifications, we introduce a graduated fitness function that tests each individual for increasing levels of skill. For example, before evolving teams are ....

John R. Koza, Forrest H Bennett III, David Andre, Martin A. Keane, and Frank Dunlap. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation, 1(2):109--128, July 1997.

....bandpass, crossover, comb, and asymmetric filters) other amplifiers, other computational circuits (square root, squaring, cubing, logarithmic, and Gaussian) a time optimal controller circuit, source identification circuits, a temperature sensing circuit, and a voltage reference circuits. See Koza, Bennett, Andre, Keane, and Dunlap 1997; Koza, Bennett, Andre, and Keane 1999; Koza, Bennett, Andre, Keane, and Brave 1999) High gain amplifiers, computational circuits, electronic thermometers, and voltage reference circuits were all covered by one or more patents when they were first invented. 9 The Illogical Nature of Creativity ....

Koza, John R., Bennett III, Forrest H, Andre, David, Keane, Martin A, and Dunlap, Frank. 1997. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation. 1(2). Pages 109 -- 128.

....solves a problem that is widely recognized as being difficult by practitioners in the field. Problem 5 in this paper and [30] 11 Automatic synthesis of a voltage reference circuit The result solves a problem that is widely recognized as being difficult by practitioners in the field. [28] 12 Automatic creation of a cellular automaton rule for the majority classification problem that is better than the Gacs Kurdyumov Levin (GKL) rule and better than all other known rules written by humans over the past 20 years The result is equal to or better than a recognized ....

....performed by designers with a large portfolio of skills. It is therefore considered by many to be a form of art rather than a science. There has been extensive previous work on the problem of circuit design using simulated annealing, artificial intelligence, and other techniques (as outlined in [28]) including work using genetic algorithms ( 36] 9] 43] However, there has previously been no general automated technique for synthesizing, from scratch, an analog electrical circuit from a high level statement of the desired behavior of the circuit. This paper presents a uniform approach ....

[Article contains additional citation context not shown here]

Koza, John R., Bennett III, Forrest H, Andre, David, Keane, Martin A, and Dunlap, Frank. 1997. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation. 1(2). Pages 109 -- 128.

No context found.

Forthcoming. Koza, John R., Bennett III, Forrest H, Andre, David, Keane, Martin A, and Dunlap, Frank. 1997. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation. 1(2). Pages 109 -- 128.

....of such subprograms, the number of arguments that each possesses, and the nature of the hierarchical references, if any, among such automatically defined functions. For current research in genetic programming, see Kinnear 1994, Angeline and Kinnear 1996, Koza, Goldberg, Fogel, and Riolo 1996, Koza et al. 1997, and Banzhaf, Nordin, Keller, and Francone 1997. A computer program is not a design. Genetic programming can be applied to circuits if a mapping is established between the program trees (rooted, point labeled trees that is, acyclic graphs with ordered branches) used in genetic programming ....

Koza, John R., Bennett III, Forrest H, Andre, David, Keane, Martin A, and Dunlap, Frank. 1997. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation. 1(2).

No context found.

J. R. Koza, F. H. Bennet III, D. Andre, M. A. Keane, and F. Dunlap, "Automated synthesis of analog electrical circuits by means of genetic programming," IEEE Transactions on Evolutionary Computation, vol. 1, no. 2, pp. 109--128, 1997.

No context found.

J. Koza, F. Bennett, D. Andre, M. Keane and F. Dunlap, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming", IEEE Trans. on Evolutionary Computation, vol. 1, no. 2, pp. 109-128, 1997.

No context found.

J. Koza et al, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming", IEEE Transactions on Evolutionary Computation, Vol.1, No.2, pp.109-128, July 1997.

No context found.

J.R. Koza, F.H. Bennett, D. Andre, M.A. Keane, F. Dunlap, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming, " IEEE Trans. on Evolutionary Computation, vol. 1, no. 2, July, 1997, pp. 109--128.

No context found.

Koza,, J.R., Bennett, F.H., Andre, D., Keane, M.A., and Dunlap, F., 1997, "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Trans. Evol. Computation, 1(2), pp.109-128.

No context found.

Koza, J.R., Bennett, F.H., Andre, D., Keane, M.A., Dunlap, F. (1997), "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Trans. on Evolutionary Computation, vol. 1, no. 2, July, pp. 109--128.

No context found.

J.R. Koza et al. Automated synthesis of analog electrical circuits by means of genetic programming. IEEE Transactions on Evolutionary Computation, 1(2):109{ 128, July 1997.

No context found.

Koza, J. R., Bennett III, F. H., Andre, D., Keane, M. A., Dunlap, F., "Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Transactions on Evolutionary Computation, Vol. 1, NO. 2, July 1997.

No context found.

J.R. Koza, F.H. Bennett, D. Andre, M.A. Keane, F. Dunlap, #Automated Synthesis of Analog Electrical Circuits by Means of Genetic Programming," IEEE Trans. on Evolutionary Computation,vol. 1, no. 2, July, 1997, pp. 109#128.

First 50 documents

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