In this paper an alternative method to symbolic segmentation is studied. Semantic segmentation be... more In this paper an alternative method to symbolic segmentation is studied. Semantic segmentation being one of the most difficult tasks currently in the computer vision area, and large number of algorithms is being developed. Thus the proposed approach in this paper exploits this large amount of available computational tools by using the algorithm selection approach. That is, let there be a set A of available algorithms for symbolic segmentation, a set of input features F, a set of image attribute A and a selection mechanism S(F, A, A) that selects on a case by case basis the best algorithm. The semantic segmentation is then an optimization process that combines best component segments from multiple results into a single optimal result. The experiments compare three different algorithm selection mechanisms using three selected semantic segmentation algorithms. The results show that using the current state of art algorithms and relatively low accuracy of algorithm selection the accuracy...
In this paper we present Quanrum Encoded Quantum Evolutionary Algorithm (QEQEA) and compare its p... more In this paper we present Quanrum Encoded Quantum Evolutionary Algorithm (QEQEA) and compare its performance against a a classical GPU accelerated Genetic Algorithm (GPUGA). The proposed QEQEA differs from existing quantum evolutionary algorithms in several points: representation of candidates circuits is using qubits and qutrits and the proposed evolutionary operators can theoretically be implemented on quantum computer provided a classical control exists. The synthesized circuits are obtained by a set of measurements performed on the encoding units of quantum representation. Both algorithms are accelerated using (general purpose graphic processing unit) GPGPU. The main target of this paper is not to propose a completely novel quantum genetic algorithm but to rather experimentally estimate the advantages of certain components of genetic algorithm being encoded and implemented in a quantum compatible manner. The algorithms are compared and evaluated on several reversible and quantum ...
Level-generalization' 88 3.6.2 Insertion principle 3.6.3 The Divide and Conquer Principle 3.6.4 T... more Level-generalization' 88 3.6.2 Insertion principle 3.6.3 The Divide and Conquer Principle 3.6.4 The Gate-collapsing principle 3.7 Examples of circuits obtained automatically for NMR technology using methods from sections 3.5 and 3.6 3.8 Chapter Conclusion 4 Genetic Algorithm for Logic Synthesis 4.1 Introduction 4.2 Genetic algorithm 4.2.1 Encoding/Representation 4.2.2 Initialization steps of GA 4.3 Evaluation of Synthesis Errors Ill 4.3.1 Element Error Evaluation method (EE) 4.3.2 Measurement Evaluation method (ME) 4.3.3 Comparison of EE and ME CONTENTS vi 4.4 Fitness functions of the GA 4.4.1 Simple Fitness Functions 4.4.2 Cost Based Fitness Functions 4.5 The Selection Process 4.6 Crossover and Mutation 4.6.1 Mutation 4.6.2 Additional GA tuning strategies 4.7 Chapter Conclusion 5 Evolutionary Search for Logic synthesis 5.1 Evolutionary Search 5.2 Experimental setup of GA 5.2.1 Input Sets of Quantum Primitives 5.3 Discussion of the Results of the Evolutionary Quantum Logic Synthesis using ME evaluation 5.3.1 Toffoli Gate 5.3.2 Synthesis of Fredkin Gate 5.3.3 Synthesis of Majority Gate 5.4 Discussion of the Results of the Evolutionary Quantum Logic Synthesis using EE evaluation CONTENTS vii 5.4.1 Toffoli Gate 5.4.2 Predkin Gate 176 5.4.3 Entanglement Circuit 5.5 Discussion of the results of the Evolutionary Synthesis .... 180 5.5.1 Results Comparison 180 5.5.2 Encountered problems during the Evolutionary QLS. . 5.6 Conclusion of the Evolutionary QLS 6 Structure Based Search for Universal Quantum Circuits 6.1 Introduction 6.2 The EX algorithm 6.3 Heuristic Search for Lowest Cost Function Class using the EX algorithm 6.3.1 New Quantum Logic Family 6.4 Discussion and Conclusion to the Structure Based Exhaustive search 7 Learning Quantum Behaviors 206 7.1 Quantum Behaviors 7.2 The concept of learning robotic behaviors from examples.. .. CONTENTS viu 7.2.1 Symbolic Quantum Synthesis of Single Output Quantum Circuits 7.3 Experiments and Results for learning Quantum Behaviors. . 7.3.1 Symbolic synthesis-Single output functions 7.4 Discussion on learning Quantum Benchmarks 7.5 Measurement Synthesis 7.5.1 Conclusions on the Measurement Dependent Quantum Logic
In this paper we present an alternative approach to symbolic segmentation; instead of implementin... more In this paper we present an alternative approach to symbolic segmentation; instead of implementing a new method we approach symbolic segmentation as an algorithm selection problem. That is, let there be n available algorithms for symbolic segmentation, a selection mechanism forms a set of input features and image attributes and selects on a case by case basis the best algorithm. The selection mechanism is demonstrated from within an algorithm framework where the selection is done in a set of various algorithm networks. Two sets of experiments are performed and in both cases we demonstrate that the algorithm selection allows to increase the result of the symbolic segmentation by a considerable amount.
35th International Symposium on Multiple-Valued Logic (ISMVL'05), 2005
The paper presents a new application of decomposition of multiple-valued relations. We developed ... more The paper presents a new application of decomposition of multiple-valued relations. We developed a theatre of interactive humanoid robots, Hahoe KAIST Robot Theatre. Version 2 includes three full body robots, equipped with vision, speech recognition, speech synthesis and natural language dialog based on machine learning abilities. The needs for this kind of project result from several research questions, especially in emotional computing and gesture generation, but the project has also educational, artistic, and entertainment values. It is a testbed to verify and integrate several algorithms in the domain of Computational Intelligence. Machine learning methods based on multiple-valued logic are used for representation of knowledge and machine learning from examples.
2013 IEEE 43rd International Symposium on Multiple-Valued Logic, 2013
We present an analysis of the Reversible and Quantum Finite State Machines (QFSM) realized as Qua... more We present an analysis of the Reversible and Quantum Finite State Machines (QFSM) realized as Quantum Circuits using the three well known sequences applied in the analysis of the classical Finite State Machines (FSM). The synchronizing, the homing and the distinguishing sequences are applied to both strictly Reversible FSM (RFSM) and QFSM in order to determine the power of these new techniques in the above mentioned new models of sequential devices. In particular, care is taken to demonstrate these classical techniques on the RFSM and the one-way QFSMs. We show certain properties of the RFSM/QFSM with respect to these sequences and we show what are the restrictions and advantages.
In real world images, many algorithms for adaptive contours detection exist and various improveme... more In real world images, many algorithms for adaptive contours detection exist and various improvements to the contours detection have been proposed. The reason for such diversity is that real world images contains heterogeneous mixtures of features and each of the available algorithms exploits some of these features. Thus, depending on the image, different algorithms shows different quality of result. In this paper we propose a method that improves the result adaptive contours detection by using an algorithm selection approach. Previous methods using the algorithm selection approach have been focusing only on images with a particular class of features (artificial, cellular) because of the complexity of real world images. In order to successfully solve this problem we first determine a set of distinctive features of each algorithm using machine learning. Then using these distinctive features we teach an algorithm selector to select best algorithm when a set of features is provided. Finally, we propose a method to split the input image into sub regions that are selected in such a manner that improves the quality of the image processing result. The proposed algorithm is verified on the set of benchmarks and its performance is comparable and better in many cases than the currently best contour detection algorithms.
Facta universitatis - series: Electronics and Energetics, 2011
We present a novel approach to the synthesis of incompletely specified reversible logic functions... more We present a novel approach to the synthesis of incompletely specified reversible logic functions. The method is based on cube grouping; the first step of the synthesis method analyzes the logic function and generates groupings of same cubes in such a manner that multiple sub-functions are realized by a single Toffoli gate. This process also reorders the function in such a manner that not only groups of similarly defined cubes are joined together but also don't care cubes. The proposed method is verified on standard benchmarks for both reversible and irreversible logic functions. The obtained results show that for functions with a significant portion of don't cares the proposed method outperforms previously proposed synthesis methods.
2013 International Joint Conference on Awareness Science and Technology & Ubi-Media Computing (iCAST 2013 & UMEDIA 2013), 2013
ABSTRACT In order to obtain the best result in image understanding it is desirable to select the ... more ABSTRACT In order to obtain the best result in image understanding it is desirable to select the best algorithm on a case by case basis. An algorithm can be selected using only image features, however such selected algorithms will often generate errors due to occlusion, shadows and other environmental conditions. To avoid such errors, it is necessary to understand processing errors on a symbolic level. Using symbolic information to determine the best algorithm is however difficult task because the possible combinations of elements and environmental conditions are almost infinite. Consequently it is impossible to predict best algorithm for all possible combinations of objects, environment conditions and context variations. In this paper we investigate selection of algorithms using symbolic image description and the determination of algorithms' error from high level image description. The proposed method transforms and minimize the high level information contained in the symbolic image description in such manner that will preserve the algorithm selection quality. The transformation takes a high level information label and transforms it into a set of generic features while the minimization uses hierarchy to reduce the specific nature of the information. Both methods of information reduction are used in a Bayesian Network because a BN is well known for using the generalization and hierarchy. As is shown in this paper, such representation efficiently reduces the fine grain high-level symbolic description to a coarser-grain hierarchy that preserves the selection quality but reduces the number of nodes.
2012 12th IEEE International Conference on Nanotechnology (IEEE-NANO), 2012
ABSTRACT We provide several extensions of the new approach to the minimization of reversible circ... more ABSTRACT We provide several extensions of the new approach to the minimization of reversible circuits based on PSE gates and ESOPOS circuits. These circuits realize the Exclusive-Or-Sum-of-Product-Sums (ESOPOS) structure where every output is an exclusive-or of Product-Sum-Exor (PSE) gates which generalize the multi-input Toffoli gates. We also propose a new efficient realization of the PSE gate that uses external-binary, internal-ternary logic.
2011 41st IEEE International Symposium on Multiple-Valued Logic, 2011
... Also, in the presented form it is necessary to keep the two transmission gates T4 and T5 so t... more ... Also, in the presented form it is necessary to keep the two transmission gates T4 and T5 so that the two opposite outputs do not generate spurious feedback signals. ... Driving fully-adiabatic logic circuits using custom high-q mems resonators. ...
2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence), 2008
Abstract— In this paper we present an evolutionary approach to the quantum symbolic logic synthes... more Abstract— In this paper we present an evolutionary approach to the quantum symbolic logic synthesis that was introduced in [1]. We use a Genetic Algorithm to synthesize quantum circuits from examples, allowing to synthesize functions that are both completely and ...
In this paper an alternative method to symbolic segmentation is studied. Semantic segmentation be... more In this paper an alternative method to symbolic segmentation is studied. Semantic segmentation being one of the most difficult tasks currently in the computer vision area, and large number of algorithms is being developed. Thus the proposed approach in this paper exploits this large amount of available computational tools by using the algorithm selection approach. That is, let there be a set A of available algorithms for symbolic segmentation, a set of input features F, a set of image attribute A and a selection mechanism S(F, A, A) that selects on a case by case basis the best algorithm. The semantic segmentation is then an optimization process that combines best component segments from multiple results into a single optimal result. The experiments compare three different algorithm selection mechanisms using three selected semantic segmentation algorithms. The results show that using the current state of art algorithms and relatively low accuracy of algorithm selection the accuracy...
In this paper we present Quanrum Encoded Quantum Evolutionary Algorithm (QEQEA) and compare its p... more In this paper we present Quanrum Encoded Quantum Evolutionary Algorithm (QEQEA) and compare its performance against a a classical GPU accelerated Genetic Algorithm (GPUGA). The proposed QEQEA differs from existing quantum evolutionary algorithms in several points: representation of candidates circuits is using qubits and qutrits and the proposed evolutionary operators can theoretically be implemented on quantum computer provided a classical control exists. The synthesized circuits are obtained by a set of measurements performed on the encoding units of quantum representation. Both algorithms are accelerated using (general purpose graphic processing unit) GPGPU. The main target of this paper is not to propose a completely novel quantum genetic algorithm but to rather experimentally estimate the advantages of certain components of genetic algorithm being encoded and implemented in a quantum compatible manner. The algorithms are compared and evaluated on several reversible and quantum ...
Level-generalization' 88 3.6.2 Insertion principle 3.6.3 The Divide and Conquer Principle 3.6.4 T... more Level-generalization' 88 3.6.2 Insertion principle 3.6.3 The Divide and Conquer Principle 3.6.4 The Gate-collapsing principle 3.7 Examples of circuits obtained automatically for NMR technology using methods from sections 3.5 and 3.6 3.8 Chapter Conclusion 4 Genetic Algorithm for Logic Synthesis 4.1 Introduction 4.2 Genetic algorithm 4.2.1 Encoding/Representation 4.2.2 Initialization steps of GA 4.3 Evaluation of Synthesis Errors Ill 4.3.1 Element Error Evaluation method (EE) 4.3.2 Measurement Evaluation method (ME) 4.3.3 Comparison of EE and ME CONTENTS vi 4.4 Fitness functions of the GA 4.4.1 Simple Fitness Functions 4.4.2 Cost Based Fitness Functions 4.5 The Selection Process 4.6 Crossover and Mutation 4.6.1 Mutation 4.6.2 Additional GA tuning strategies 4.7 Chapter Conclusion 5 Evolutionary Search for Logic synthesis 5.1 Evolutionary Search 5.2 Experimental setup of GA 5.2.1 Input Sets of Quantum Primitives 5.3 Discussion of the Results of the Evolutionary Quantum Logic Synthesis using ME evaluation 5.3.1 Toffoli Gate 5.3.2 Synthesis of Fredkin Gate 5.3.3 Synthesis of Majority Gate 5.4 Discussion of the Results of the Evolutionary Quantum Logic Synthesis using EE evaluation CONTENTS vii 5.4.1 Toffoli Gate 5.4.2 Predkin Gate 176 5.4.3 Entanglement Circuit 5.5 Discussion of the results of the Evolutionary Synthesis .... 180 5.5.1 Results Comparison 180 5.5.2 Encountered problems during the Evolutionary QLS. . 5.6 Conclusion of the Evolutionary QLS 6 Structure Based Search for Universal Quantum Circuits 6.1 Introduction 6.2 The EX algorithm 6.3 Heuristic Search for Lowest Cost Function Class using the EX algorithm 6.3.1 New Quantum Logic Family 6.4 Discussion and Conclusion to the Structure Based Exhaustive search 7 Learning Quantum Behaviors 206 7.1 Quantum Behaviors 7.2 The concept of learning robotic behaviors from examples.. .. CONTENTS viu 7.2.1 Symbolic Quantum Synthesis of Single Output Quantum Circuits 7.3 Experiments and Results for learning Quantum Behaviors. . 7.3.1 Symbolic synthesis-Single output functions 7.4 Discussion on learning Quantum Benchmarks 7.5 Measurement Synthesis 7.5.1 Conclusions on the Measurement Dependent Quantum Logic
In this paper we present an alternative approach to symbolic segmentation; instead of implementin... more In this paper we present an alternative approach to symbolic segmentation; instead of implementing a new method we approach symbolic segmentation as an algorithm selection problem. That is, let there be n available algorithms for symbolic segmentation, a selection mechanism forms a set of input features and image attributes and selects on a case by case basis the best algorithm. The selection mechanism is demonstrated from within an algorithm framework where the selection is done in a set of various algorithm networks. Two sets of experiments are performed and in both cases we demonstrate that the algorithm selection allows to increase the result of the symbolic segmentation by a considerable amount.
35th International Symposium on Multiple-Valued Logic (ISMVL'05), 2005
The paper presents a new application of decomposition of multiple-valued relations. We developed ... more The paper presents a new application of decomposition of multiple-valued relations. We developed a theatre of interactive humanoid robots, Hahoe KAIST Robot Theatre. Version 2 includes three full body robots, equipped with vision, speech recognition, speech synthesis and natural language dialog based on machine learning abilities. The needs for this kind of project result from several research questions, especially in emotional computing and gesture generation, but the project has also educational, artistic, and entertainment values. It is a testbed to verify and integrate several algorithms in the domain of Computational Intelligence. Machine learning methods based on multiple-valued logic are used for representation of knowledge and machine learning from examples.
2013 IEEE 43rd International Symposium on Multiple-Valued Logic, 2013
We present an analysis of the Reversible and Quantum Finite State Machines (QFSM) realized as Qua... more We present an analysis of the Reversible and Quantum Finite State Machines (QFSM) realized as Quantum Circuits using the three well known sequences applied in the analysis of the classical Finite State Machines (FSM). The synchronizing, the homing and the distinguishing sequences are applied to both strictly Reversible FSM (RFSM) and QFSM in order to determine the power of these new techniques in the above mentioned new models of sequential devices. In particular, care is taken to demonstrate these classical techniques on the RFSM and the one-way QFSMs. We show certain properties of the RFSM/QFSM with respect to these sequences and we show what are the restrictions and advantages.
In real world images, many algorithms for adaptive contours detection exist and various improveme... more In real world images, many algorithms for adaptive contours detection exist and various improvements to the contours detection have been proposed. The reason for such diversity is that real world images contains heterogeneous mixtures of features and each of the available algorithms exploits some of these features. Thus, depending on the image, different algorithms shows different quality of result. In this paper we propose a method that improves the result adaptive contours detection by using an algorithm selection approach. Previous methods using the algorithm selection approach have been focusing only on images with a particular class of features (artificial, cellular) because of the complexity of real world images. In order to successfully solve this problem we first determine a set of distinctive features of each algorithm using machine learning. Then using these distinctive features we teach an algorithm selector to select best algorithm when a set of features is provided. Finally, we propose a method to split the input image into sub regions that are selected in such a manner that improves the quality of the image processing result. The proposed algorithm is verified on the set of benchmarks and its performance is comparable and better in many cases than the currently best contour detection algorithms.
Facta universitatis - series: Electronics and Energetics, 2011
We present a novel approach to the synthesis of incompletely specified reversible logic functions... more We present a novel approach to the synthesis of incompletely specified reversible logic functions. The method is based on cube grouping; the first step of the synthesis method analyzes the logic function and generates groupings of same cubes in such a manner that multiple sub-functions are realized by a single Toffoli gate. This process also reorders the function in such a manner that not only groups of similarly defined cubes are joined together but also don't care cubes. The proposed method is verified on standard benchmarks for both reversible and irreversible logic functions. The obtained results show that for functions with a significant portion of don't cares the proposed method outperforms previously proposed synthesis methods.
2013 International Joint Conference on Awareness Science and Technology & Ubi-Media Computing (iCAST 2013 & UMEDIA 2013), 2013
ABSTRACT In order to obtain the best result in image understanding it is desirable to select the ... more ABSTRACT In order to obtain the best result in image understanding it is desirable to select the best algorithm on a case by case basis. An algorithm can be selected using only image features, however such selected algorithms will often generate errors due to occlusion, shadows and other environmental conditions. To avoid such errors, it is necessary to understand processing errors on a symbolic level. Using symbolic information to determine the best algorithm is however difficult task because the possible combinations of elements and environmental conditions are almost infinite. Consequently it is impossible to predict best algorithm for all possible combinations of objects, environment conditions and context variations. In this paper we investigate selection of algorithms using symbolic image description and the determination of algorithms' error from high level image description. The proposed method transforms and minimize the high level information contained in the symbolic image description in such manner that will preserve the algorithm selection quality. The transformation takes a high level information label and transforms it into a set of generic features while the minimization uses hierarchy to reduce the specific nature of the information. Both methods of information reduction are used in a Bayesian Network because a BN is well known for using the generalization and hierarchy. As is shown in this paper, such representation efficiently reduces the fine grain high-level symbolic description to a coarser-grain hierarchy that preserves the selection quality but reduces the number of nodes.
2012 12th IEEE International Conference on Nanotechnology (IEEE-NANO), 2012
ABSTRACT We provide several extensions of the new approach to the minimization of reversible circ... more ABSTRACT We provide several extensions of the new approach to the minimization of reversible circuits based on PSE gates and ESOPOS circuits. These circuits realize the Exclusive-Or-Sum-of-Product-Sums (ESOPOS) structure where every output is an exclusive-or of Product-Sum-Exor (PSE) gates which generalize the multi-input Toffoli gates. We also propose a new efficient realization of the PSE gate that uses external-binary, internal-ternary logic.
2011 41st IEEE International Symposium on Multiple-Valued Logic, 2011
... Also, in the presented form it is necessary to keep the two transmission gates T4 and T5 so t... more ... Also, in the presented form it is necessary to keep the two transmission gates T4 and T5 so that the two opposite outputs do not generate spurious feedback signals. ... Driving fully-adiabatic logic circuits using custom high-q mems resonators. ...
2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence), 2008
Abstract— In this paper we present an evolutionary approach to the quantum symbolic logic synthes... more Abstract— In this paper we present an evolutionary approach to the quantum symbolic logic synthesis that was introduced in [1]. We use a Genetic Algorithm to synthesize quantum circuits from examples, allowing to synthesize functions that are both completely and ...
Uploads
Papers by Martin Lukac