Recent scientific writings and speeches

  1. GPU-based Tuning of Quantum-Inspired Genetic Algorithm for a Combinatorial Optimization Problem
    R.Nowotniak, J. Kucharski
    Abstract: This paper concerns efficient parameters tuning (meta-optimization) of a state-of-the-art metaheuristic, Quantum-Inspired Genetic Algorithm (QIGA), in a GPU-based massively parallel computing environment (NVidia CUDATMtechnology). A novel approach to parallel implementation of the algorithm has been presented. In a block of threads, each thread transforms a separate quantum individual or different quantum gene; In each block, a separate experiment with different population is conducted. The computations have been distributed to eight GPU devices, and over 400x speedup has been gained in comparison to Intel Core i7 2.93GHz CPU. This approach allows efficient meta-optimization of the algorithm parameters. Two criteria for the meta-optimization of the rotation angles in quantum genes state space have been considered. Performance comparison has been performed on combinatorial optimization (knapsack problem), and it has been presented that the tuned algorithm is superior to Simple Genetic Algorithm and to original QIGA algorithm.
    Published in Bulletin of The Polish Academy of Sciences: TechnicalSciences,
    Vol. 60, No. 2, 2012, ISSN 0239-7528
    paper (PDF) in the Bulletin
  2. Meta-optimization of Quantum-Inspired Evolutionary Algorithms in The Polish Grid Infrastructure
    R.Nowotniak
    Abstract:The presented research concerns meta-optimization of Quantum-Inspired Evolutionary Algorithms (QIEA) with the use of computational resources of The Polish Grid Infrastructure PL-GRID. In QIEA algorithms, additional elements of randomness are introduced. They are inspired by properties of quantum mechanical systems. The intensified random factors strongly influence effectiveness of the algorithms in numerical and combinatorial optimization problems. Parameters of QIEA algorithms determine the influence of the random factors. Automatic tuning of the parameters values is possible by state-of-the-art meta-optimization techniques, which involves a considerable computational cost.
    Published in Proceedings of the 2nd Scientific Session of TUL PhD Students, ISBN 978-83-7283-490-4
    The Session was held in April 2012 in Rogow, Poland
    abstract from the Scientific Session booklet
  3. GPU-based Tuning of Quantum-Inspired Genetic Algorithm for a Combinatorial Optimization Problem
    R.Nowotniak, J. Kucharski
    Abstract: This paper concerns efficient parameters tuning (meta-optimization) of a state-of-the-art metaheuristic, Quantum-Inspired Genetic Algorithm, in a GPU-based massively parallel computing environment (NVidia CUDA™ technology). A novel approach to parallel implementation of the algorithm has been presented. The computations have been distributed to eight GPU devices, and over 400x speedup has been gained. This approach allows efficient meta-optimization of the algorithm which has been demonstrated on combinatorial optimization (knapsack problem).
    Published in Proceedings of the XIV International Conference SYSTEM MODELING and CONTROL
    JUNE 27-29, 2011, ISSN 978-83-927875-1-8
    paper (PDF) in the conference proceedings
    presentation (PDF) from the conference
  4. GPU-based massively parallel implementation of metaheuristic algorithms
    R.Nowotniak, J. Kucharski
    Abstract: In this paper, implementation of Quantum-Inspired Genetic Algorithm(QIGA) in massively parallel environment (Graphics Processing Units) has been presented. Contrary to many recent papers concerning parallel implementation of evolutionary algorithms, in this paper a novel approach has been taken. QIGA algorithm has been implemented entirely as a computational kernel. Parallelization of the algorithm has been performed on two levels: In a block of threads, each thread transforms a separate individual or different gene; In each block, separate populations with same or different parameters are evolved. Finally, the computations have been distributed to eight GPU devices, and over 400x speedup has been gained in comparison to sequential implementation of the algorithm in ANSI C on one Intel Core i7 2.93GHz CPU core. Correctness of the results has been verified in statistical analysis. The presented approach can be applied to experimentation with a broad class of metaheuristics.
    The Conference was held in June 2011 in Słok, Poland
    paper
     (PDF) in the conference proceedings
    presentation and poster from the conference
  5. Modelling Reality In Visual Python
    R.Nowotniak, C.Draus, M. Nowak, G.Rybak
    Abstract: In summer semester of the academic year 2010/2011, the course entitled “Programming in Scripting Languages” was given in Computer Engineering Department. In the course laboratories, Python[2] programming language was used, and the attending students have created various projects in Visual Python. The students projects were created in partial fulfilment of the requirements for receiving a credit for the laboratories. Based on the students projects, the present paper concerns modelling reality, visualizing real-world phenomena and writing simple educational games in Visual Python.
    Published in Proceedings of the INotice 2011 Conference, ISBN 978-83-7283-407-2
    The Conference was held in June 2011 in Słok, Poland
    paper (PDF) in the conference proceedings
    presentation and movie from the conference
  6. Quantum-Inspired Evolutionary Algorithms in Search and Optimization
    R.Nowotniak
    Abstract: The presented research concerns a new class of artificial intelligence techniques, drawing inspiration from both the biological evolution and unitary evolution of quantum systems. Intensive studies in this new area (Quantum-Inspired Evolutionary Algorithms) have been conducted since the beginning of the last decade. Quantum computing provides a valuable source of inspiration to create new heuristic methods for searching and optimization. Additional random factors, inspired by the unique quantum phenomena (superposition of states, quantum parallelism, interference, probability amplitudes), introduce a “new dimension” to the evolutionary algorithms.  […]
    Published in Proceedings of the Scientific Session of TUL PhD Students, ISBN 978-83-7283-411-9
    The Session was held in April 2011 in Rogow, Poland
    abstract (PDF) from the Scientific Session booklet
    my presentation (PDF) from the Session
  7. Meta-optimization of Quantum-Inspired Evolutionary Algorithm
    R.Nowotniak, J. Kucharski
    Abstract: In this paper, a meta-optimization algorithm, based on Local Unimodal Sampling (LUS), has been applied to tune selected parameters of Quantum-Inspired Evolutionary Algorithm for numerical optimization problems coded in real numbers. Tuning of the following two parameters has been considered: crossover rate and contraction factor. Performance landscapes of the algorithm meta-fitness have been approximated and numerical experiments have been conducted on four high-dimensional optimization problems. The tuned algorithm outperformed the algorithm with initially proposed parameters in the conducted experiments.
    Published in Proceedings of the XVII International Conference on Information Technology Systems,
    ISBN 978-83-7283-378-5
    paper (PDF) in the conference proceedings
    presentation (and two animations) and screencast from the conference
  8. Building Blocks Propagation in Quantum-Inspired Genetic Algorithm
    R.Nowotniak, J. Kucharski
    Abstract: This paper presents an analysis of building blocks propagation in Quantum-Inspired Genetic Algorithm, which belongs to a new class of metaheuristics drawing their inspiration from both the biological evolution and unitary evolution of quantum systems. The expected number of quantum chromosomes matching a schema has been analysed and a random variable corresponding to this issue has been introduced. The results have been compared with Simple Genetic Algorithm. Also, it has been presented how selected binary quantum chromosomes cover a domain of one-dimensional fitness function.
    preprint (PDF) from Scientific Bulletin of Academy of Science and Technology, Automatics, 2010
    presentation and poster on SŁOK 2010 conference
  9. Survey of Quantum-Inspired Evolutionary Algorithms
    R.Nowotniak
    Abstract: This paper presents a concise survey of a new class of metaheuristics, drawing their inspiration from both: biological evolution and unitary evolution of quantum systems. In the first part of the paper, general concepts behind quantum-inspired evolutionary algorithms have been presented. In the second part, a state of the art of this field has been presented and a literature review has been conducted.
    Published in Proceedings of the FIMB PhD students conference
    ISSN 2082-4831
    paper (PDF ) in the conference proceedings
    presentation from the conference
  10. Comparison of Algorithms for Simultaneous Localization and Mapping Problem for Mobile Robot
    S.Jeżewski, M. Łaski, R. Nowotniak
    Abstract: This paper presents a comparison of selected algorithms for simultaneous localization and mapping (SLAM) problem in mobile robotics. Results of four general metaheuristics, Simple Genetic Algorithm, Particle Swarm Optimization, Quantum-Inspired Genetic Algorithms and Genetic Algorithm with Quantum Probability Representation, have been compared to results of classical, analytic method in this field, Iterative Closes Points algorithm. In the experiments the same objective function, drawn from Iterative Closest Points algorithm, has been used. Two situations have been considered: local and global localization problems of mobile robot. Both problems are import and often critical for successful navigation of robot in environment.
    preprint (PDF) from Scientific Bulletin of Academy of Science and Technology, Automatics, 2010
    poster from SŁOK 2010 conference
  11. Quantum-Inspired Evolutionary Algorithms
    R. Nowotniak
    my 5 minutes-long presentation from a PhD students’ seminar at Computer Engineering Department
    January 20, 2010
  12. Application of Quantum Genetic Algorithms in Feature Selection Problem
    Ł. Jopek, R. Nowotniak, M. Postolski, L. Babout, M. Janaszewski
    Abstract: In the article, a feature selection problem for k-NN classifier in image segmentation has been analyzed. Feature selection has been considered as a two criteria combinatorial optimization problem. An objective of optimization process was to find an image points’ features subset, allowing good quality of segmentation in satisfactory time. A fitness function for features subsets has been proposed, taking into account time of features calculation and quality of segmentation. Three population-based heuristic methods of optimization have been compared: simple genetic algorithm and its two modifications, inspired by principles of quantum computing: QiGA (Quantum-Inspired Genetic Algorithm) and GAQPR (Genetic Algorithm with Quantum Probability Representation). Results of experiments with artificial and tomography textures have been presented.
    Published in Scientific Bulletin of Academy of Science and Technology, Automatics, 2009
    ISSN 1429-3447
    paper (PDF, in Polish)
    presentation and poster on SŁOK 2009 conference
    June 24, 2009
  13. On The Evolutionary Design of Quantum Algorithms
    R. Nowotniak
    presentation
    on Computer Engineering Department seminar
    text of my speech (in Polish )
    poster (in Polish ) on XVI International Conference on Information Technology Systems
    October 1, 2008
  14. On The Evolutionary Design of Quantum Algorithm
    R. Nowotniak
    Abstract
    : This master thesis is focused on quantum computing and its possible applications of genetic algorithms. Quantum computing is considered in the terms of quantum gates computational model. Proposal of object model for quantum computing is presented. Practical part of this thesis consists of the quantum computing library (qclib) allowing to simulate a quantum computer. The basic quantum algorithms are presented and implemented using qclib: Grover’s fast search algorithm, quantum teleportation, quantum superdense coding and production of entangled quantum states. Generation of such quantum algorithms is a difficult task for human. They are unintuitive and computationally intensive. For that reason hybrid systems combining quantum computing and genetic algorithms are considered. In this thesis simple and modified genetic algorithms are used and compared to evolve quantum gates. Complete quantum circuits are created using genetic programming. Some of possible approaches are illustrated with numerical experiments.
    my master’s thesis (in Polish )
    July 1, 2008
  15. Quantum Teleportation
    R. Nowotniak
    presentation (in Polish )
    Students Scientific Group seminar, Institute of Computer Science
    January 9, 2008
  16. Grover’s Quantum Search Algorithm
    R. Nowotniak
    presentation
    (in Polish )
    my presentation on Institute of Computer Science seminar
    November 13, 2007
  17. Quantum Computing
    R. Nowotniak
    presentation (in Polish ), poster (in Polish ) and text of my speech (in Polish )
    XV International Conference on Information Technology Systems
    October 26, 2007
  18. Grover’s Quantum Search Algorithm
    R. Nowotniak
    presentation (in Polish )
    Students Scientific Group seminar, Institute of Computer Science
    October 24, 2007
  19. Quantum Computing in Numerical Python
    R. Nowotniak
    presentation
    (in Polish )
    Students Scientific Group seminar, Institute of Computer Science
    October 10, 2007
  20. Quantum Cooperation
    R. Nowotniak
    presentation (in Polish )
    Students Scientific Group seminar, Institute of Computer Science
    May 17, 2007
  21. Quantum Programming Languages
    R. Nowotniak
    presentation (in Polish )
    Students Scientific Group seminar, Institute of Computer Science
    January 3, 2007