The paper explores application of genetic programming to evolution of robot morphology, and co-ev... more The paper explores application of genetic programming to evolution of robot morphology, and co-evolution of morphology and low-level control. Extensions to standard genetic programming are presented that allow for straight-forward storage, retrieval, transfer, modification of data stored in the context of a syntactic tree, and shared by multiple nodes. These extensions are used to embed a genetic algorithm within the genetic programming approach to evolve values of constants. Experimental results are presented and evaluated.
2011 IEEE International Conference on Intelligent Computing and Intelligent Systems, 2011
This paper concerns applications of genetic algorithms and genetic programming to tasks for which... more This paper concerns applications of genetic algorithms and genetic programming to tasks for which it is difficult to find a representation that does not map to a highly complex and discontinuous fitness landscape. In such cases the standard algorithm is prone to getting trapped in local extremes. The paper proposes several adaptive mechanisms that are useful in preventing the search from getting trapped.
11th IFAC/IEEE International Conference on Pro- grammable Devices and Embedded Systems, 2012
The paper presents a new approach towards modular genetic programming. The modules are generated ... more The paper presents a new approach towards modular genetic programming. The modules are generated automatically and dynamically. They are node-attached and contained in the individual. Their ancestry is being tracked so as to allow for remapping to related modules. The performance of the proposed framework is demonstrated on the even-n-parity problem. A comparison with the results achieved using Automatically Defined Functions as proposed by John Koza is provided.
The paper explores application of genetic programming to evolution of robot morphology, and co-ev... more The paper explores application of genetic programming to evolution of robot morphology, and co-evolution of morphology and low-level control. Extensions to standard genetic programming are presented that allow for straight-forward storage, retrieval, transfer, modification of data stored in the context of a syntactic tree, and shared by multiple nodes. These extensions are used to embed a genetic algorithm within the genetic programming approach to evolve values of constants. Experimental results are presented and evaluated.
Proceedings of 10th International Conference, ELEKTRO 2014, 2014
The paper elaborates upon a prior proposal for a novelty detector based on an artificial neural n... more The paper elaborates upon a prior proposal for a novelty detector based on an artificial neural network forecaster. In the former paper, the novelty-based motivational signal was used in place of more conventional techniques (such as the ε-greedy policy, or the softmax policy) to drive exploration, in the context of V-learning. The current paper provides a more comprehensive study of such handling of the exploration vs. exploitation trade-off. It also studies the various problems concerning application of the approach to SARSA, and Q-learning. Also, and with the same goal in mind, the paper presents several advances upon the original design.
Proceedings of the Conference on Modern Information Technology in the Innovation Processs of the Industrial Enterprises, 2012
This paper deals with computer simulation of manufacturing systems and on emulation as an advance... more This paper deals with computer simulation of manufacturing systems and on emulation as an advanced approach of simulation of production and logistic processes. It focuses models going out of simulation models - so called metamodels - as one of the possible ways how to increase effectiveness of simulation optimization and how to reduce the requirements of time-consuming simulation of manufacturing systems. It describes the basis of metamodelling process on specific example of ZIMS and describes neural network metamodels as an universal function approximators.
Prvý diel vysokoškolskej učebnice k predmetu Umelá inteligencia. Venuje sa dvom témam: fuzzy prís... more Prvý diel vysokoškolskej učebnice k predmetu Umelá inteligencia. Venuje sa dvom témam: fuzzy prístupom a umelým neurónovým sieťam. Postupne preberá základné koncepty z teórie fuzzy množín, fuzzy logiky a predstavuje ich aplikácie vo fuzzy riadení a fuzzy aritmetike. V ďalšej časti sa venuje dopredným neurónovým sieťam a predstavuje základné metódy kontrolovaného učenia v nich. Z metód nekontrolovaného učenia sa venuje konkurenčnému učeniu a samoorganizujúcim mapám. Na záver predstavuje aplikácie umelých neurónových sietí v automatickom riadení. Výklad je doplnený ilustráciami a príkladmi zdrojového kódu pre prostredie Matlab. V závere každej kapitoly je časť s odkazmi na ďalšie súvisiace odborné materiály a nástroje.
The paper explores application of genetic programming to evolution of robot morphology, and co-ev... more The paper explores application of genetic programming to evolution of robot morphology, and co-evolution of morphology and low-level control. Extensions to standard genetic programming are presented that allow for straight-forward storage, retrieval, transfer, modification of data stored in the context of a syntactic tree, and shared by multiple nodes. These extensions are used to embed a genetic algorithm within the genetic programming approach to evolve values of constants. Experimental results are presented and evaluated.
2011 IEEE International Conference on Intelligent Computing and Intelligent Systems, 2011
This paper concerns applications of genetic algorithms and genetic programming to tasks for which... more This paper concerns applications of genetic algorithms and genetic programming to tasks for which it is difficult to find a representation that does not map to a highly complex and discontinuous fitness landscape. In such cases the standard algorithm is prone to getting trapped in local extremes. The paper proposes several adaptive mechanisms that are useful in preventing the search from getting trapped.
11th IFAC/IEEE International Conference on Pro- grammable Devices and Embedded Systems, 2012
The paper presents a new approach towards modular genetic programming. The modules are generated ... more The paper presents a new approach towards modular genetic programming. The modules are generated automatically and dynamically. They are node-attached and contained in the individual. Their ancestry is being tracked so as to allow for remapping to related modules. The performance of the proposed framework is demonstrated on the even-n-parity problem. A comparison with the results achieved using Automatically Defined Functions as proposed by John Koza is provided.
The paper explores application of genetic programming to evolution of robot morphology, and co-ev... more The paper explores application of genetic programming to evolution of robot morphology, and co-evolution of morphology and low-level control. Extensions to standard genetic programming are presented that allow for straight-forward storage, retrieval, transfer, modification of data stored in the context of a syntactic tree, and shared by multiple nodes. These extensions are used to embed a genetic algorithm within the genetic programming approach to evolve values of constants. Experimental results are presented and evaluated.
Proceedings of 10th International Conference, ELEKTRO 2014, 2014
The paper elaborates upon a prior proposal for a novelty detector based on an artificial neural n... more The paper elaborates upon a prior proposal for a novelty detector based on an artificial neural network forecaster. In the former paper, the novelty-based motivational signal was used in place of more conventional techniques (such as the ε-greedy policy, or the softmax policy) to drive exploration, in the context of V-learning. The current paper provides a more comprehensive study of such handling of the exploration vs. exploitation trade-off. It also studies the various problems concerning application of the approach to SARSA, and Q-learning. Also, and with the same goal in mind, the paper presents several advances upon the original design.
Proceedings of the Conference on Modern Information Technology in the Innovation Processs of the Industrial Enterprises, 2012
This paper deals with computer simulation of manufacturing systems and on emulation as an advance... more This paper deals with computer simulation of manufacturing systems and on emulation as an advanced approach of simulation of production and logistic processes. It focuses models going out of simulation models - so called metamodels - as one of the possible ways how to increase effectiveness of simulation optimization and how to reduce the requirements of time-consuming simulation of manufacturing systems. It describes the basis of metamodelling process on specific example of ZIMS and describes neural network metamodels as an universal function approximators.
Prvý diel vysokoškolskej učebnice k predmetu Umelá inteligencia. Venuje sa dvom témam: fuzzy prís... more Prvý diel vysokoškolskej učebnice k predmetu Umelá inteligencia. Venuje sa dvom témam: fuzzy prístupom a umelým neurónovým sieťam. Postupne preberá základné koncepty z teórie fuzzy množín, fuzzy logiky a predstavuje ich aplikácie vo fuzzy riadení a fuzzy aritmetike. V ďalšej časti sa venuje dopredným neurónovým sieťam a predstavuje základné metódy kontrolovaného učenia v nich. Z metód nekontrolovaného učenia sa venuje konkurenčnému učeniu a samoorganizujúcim mapám. Na záver predstavuje aplikácie umelých neurónových sietí v automatickom riadení. Výklad je doplnený ilustráciami a príkladmi zdrojového kódu pre prostredie Matlab. V závere každej kapitoly je časť s odkazmi na ďalšie súvisiace odborné materiály a nástroje.
Uploads