2018 |
Liskowski, Paweł, Blądek, Iwo, Krawiec, Krzysztof Neuro-guided Genetic Programming: Prioritizing Evolutionary Search with Neural Networks (Inproceeding) Proceedings of the Genetic and Evolutionary Computation Conference, pp. 1143–1150, ACM, Kyoto, Japan, 2018, ISBN: 978-1-4503-5618-3. (Links | BibTeX) @inproceedings{Liskowski:2018:NGP:3205455.3205629,
title = {Neuro-guided Genetic Programming: Prioritizing Evolutionary Search with Neural Networks},
author = {Liskowski, Paweł and Blądek, Iwo and Krawiec, Krzysztof},
url = {http://doi.acm.org/10.1145/3205455.3205629},
isbn = {978-1-4503-5618-3},
year = {2018},
date = {2018-01-01},
booktitle = {Proceedings of the Genetic and Evolutionary Computation Conference},
pages = {1143--1150},
publisher = {ACM},
address = {Kyoto, Japan},
series = {GECCO '18},
keywords = {}
}
|
Liskowski, Paweł, Wieloch, Bartosz, Krawiec, Krzysztof Neural Estimation of Interaction Outcomes (Inproceeding) Proceedings of the Genetic and Evolutionary Computation Conference, pp. 1055–1062, ACM, Kyoto, Japan, 2018, ISBN: 978-1-4503-5618-3. (Links | BibTeX) @inproceedings{Liskowski:2018:NEI:3205455.3205600,
title = {Neural Estimation of Interaction Outcomes},
author = {Liskowski, Paweł and Wieloch, Bartosz and Krawiec, Krzysztof},
url = {http://doi.acm.org/10.1145/3205455.3205600},
isbn = {978-1-4503-5618-3},
year = {2018},
date = {2018-01-01},
booktitle = {Proceedings of the Genetic and Evolutionary Computation Conference},
pages = {1055--1062},
publisher = {ACM},
address = {Kyoto, Japan},
series = {GECCO '18},
keywords = {}
}
|
Paweł Liskowski Heuristic Algorithms for Discovery of Search Objectives in Test-based Problems (PhD Thesis) Poznań University of Technology, 2018. (Links | BibTeX) @phdthesis{LiskowskiPhd2018,
title = {Heuristic Algorithms for Discovery of Search Objectives in Test-based Problems},
author = {Paweł Liskowski},
url = {http://www.cs.put.poznan.pl/pliskowski/pub/phdthesis.pdf},
year = {2018},
date = {2018-01-01},
address = {Poznan, Poland},
school = {Poznań University of Technology},
keywords = {}
}
|
2017 |
Paweł Liskowski, Wojciech Jaśkowski Accelerating Coevolution with Adaptive Matrix Factorization (Inproceeding) Proceedings of the Genetic and Evolutionary Computation Conference, pp. 457–464, ACM, Berlin, Germany, 2017, ISBN: 978-1-4503-4920-8. (Links | BibTeX) @inproceedings{Liskowski:2017:ACA:3071178.3071320,
title = {Accelerating Coevolution with Adaptive Matrix Factorization},
author = {Paweł Liskowski and Wojciech Jaśkowski},
url = {http://doi.acm.org/10.1145/3071178.3071320},
isbn = {978-1-4503-4920-8},
year = {2017},
date = {2017-01-01},
booktitle = {Proceedings of the Genetic and Evolutionary Computation Conference},
pages = {457--464},
publisher = {ACM},
address = {Berlin, Germany},
series = {GECCO '17},
keywords = {}
}
|
Paweł Liskowski, Krzysztof Krawiec Discovery of Search Objectives in Continuous Domains (Inproceeding) Proceedings of the Genetic and Evolutionary Computation Conference, pp. 969–976, ACM, Berlin, Germany, 2017, ISBN: 978-1-4503-4920-8. (Links | BibTeX) @inproceedings{Liskowski:2017:DSO:3071178.3071344,
title = {Discovery of Search Objectives in Continuous Domains},
author = {Paweł Liskowski and Krzysztof Krawiec},
url = {http://doi.acm.org/10.1145/3071178.3071344},
isbn = {978-1-4503-4920-8},
year = {2017},
date = {2017-01-01},
booktitle = {Proceedings of the Genetic and Evolutionary Computation Conference},
pages = {969--976},
publisher = {ACM},
address = {Berlin, Germany},
series = {GECCO '17},
keywords = {}
}
|
Paweł Liskowski, Krzysztof Krawiec Adaptive Test Selection for Factorization-based Surrogate Fitness in Genetic Programming (Article) Foundations of Computing and Decision Sciences, (accepted), 2017. (BibTeX) @article{liskowski17adaptive,
title = {Adaptive Test Selection for Factorization-based Surrogate Fitness in Genetic Programming},
author = {Paweł Liskowski and Krzysztof Krawiec},
year = {2017},
date = {2017-01-01},
journal = {Foundations of Computing and Decision Sciences},
volume = {(accepted)},
keywords = {}
}
|
Paweł Liskowski, Wojciech Jaśkowski, Krzysztof Krawiec Learning to Play Othello with Deep Neural Networks (Article) IEEE Transactions on Computational Intelligence and AI in Games, (accepted), 2017. (BibTeX) @article{liskowski2017learning,
title = {Learning to Play Othello with Deep Neural Networks},
author = {Paweł Liskowski, Wojciech Jaśkowski, Krzysztof Krawiec},
year = {2017},
date = {2017-01-01},
journal = {IEEE Transactions on Computational Intelligence and AI in Games},
volume = {(accepted)},
publisher = {IEEE},
keywords = {}
}
|
Maciej Szkulmowski, Paweł Liskowski, Bartosz Wieloch, Krzysztof Krawiec, Bartosz Sikorski Convolutional neural networks for artifact free OCT retinal angiography (Article) Investigative Ophthalmology & Visual Science, 58 (8), pp. 649–649, 2017. (BibTeX) @article{szkulmowski2017convolutional,
title = {Convolutional neural networks for artifact free OCT retinal angiography},
author = {Maciej Szkulmowski and Paweł Liskowski and Bartosz Wieloch and Krzysztof Krawiec and Bartosz Sikorski},
year = {2017},
date = {2017-01-01},
journal = {Investigative Ophthalmology & Visual Science},
volume = {58},
number = {8},
pages = {649--649},
publisher = {The Association for Research in Vision and Ophthalmology},
keywords = {}
}
|
2016 |
Paweł Liskowski, Krzysztof Krawiec Online Discovery of Search Objectives for Test-Based Problems (Article) Evolutionary Computation, 25 (3), pp. 375–406, 2016. (Links | BibTeX) @article{doi:10.1162/evco_a_00179,
title = {Online Discovery of Search Objectives for Test-Based Problems},
author = {Paweł Liskowski and Krzysztof Krawiec},
url = {http://www.mitpressjournals.org/doi/abs/10.1162/evco_a_00179},
year = {2016},
date = {2016-01-01},
journal = {Evolutionary Computation},
volume = {25},
number = {3},
pages = {375--406},
keywords = {}
}
|
Paweł Liskowski, Krzysztof Krawiec Segmenting Retinal Blood Vessels With Deep Neural Networks (Article) IEEE transactions on medical imaging, 35 (11), pp. 2369–2380, 2016. (Links | BibTeX) @article{liskowski2016segmenting,
title = {Segmenting Retinal Blood Vessels With Deep Neural Networks},
author = {Paweł Liskowski and Krzysztof Krawiec},
url = {http://ieeexplore.ieee.org/document/7440871/},
year = {2016},
date = {2016-01-01},
journal = {IEEE transactions on medical imaging},
volume = {35},
number = {11},
pages = {2369--2380},
publisher = {IEEE},
keywords = {}
}
|
Paweł Liskowski, Krzysztof Krawiec Non-negative Matrix Factorization for Unsupervised Derivation of Search Objectives in Genetic Programming (Inproceeding) Proceedings of the 2016 on Genetic and Evolutionary Computation Conference, pp. 749–756, ACM, Denver, Colorado, USA, 2016, ISBN: 978-1-4503-4206-3. (Links | BibTeX) @inproceedings{Liskowski:2016:NMF:2908812.2908888,
title = {Non-negative Matrix Factorization for Unsupervised Derivation of Search Objectives in Genetic Programming},
author = {Paweł Liskowski and Krzysztof Krawiec},
url = {http://doi.acm.org/10.1145/2908812.2908888},
isbn = {978-1-4503-4206-3},
year = {2016},
date = {2016-01-01},
booktitle = {Proceedings of the 2016 on Genetic and Evolutionary Computation Conference},
pages = {749--756},
publisher = {ACM},
address = {Denver, Colorado, USA},
series = {GECCO '16},
keywords = {}
}
|
Paweł Liskowski, Krzysztof Krawiec Surrogate Fitness via Factorization of Interaction Matrix (Inproceeding) Heywood, Malcolm; McDermott, James; Castelli, Mauro; Costa, Ernesto (Ed.): EuroGP 2016: Proceedings of the 19th European Conference on Genetic Programming, pp. 65–79, Springer Verlag, Porto, Portugal, 2016. (Abstract | Links | BibTeX) @inproceedings{Liskowski:2016:EuroGP,
title = {Surrogate Fitness via Factorization of Interaction Matrix},
author = {Paweł Liskowski and Krzysztof Krawiec},
editor = {Malcolm I. Heywood and James McDermott and Mauro Castelli and Ernesto Costa},
url = {https://link.springer.com/chapter/10.1007/978-3-319-30668-1_5},
year = {2016},
date = {2016-01-01},
booktitle = {EuroGP 2016: Proceedings of the 19th European Conference on Genetic Programming},
volume = {9594},
pages = {65--79},
publisher = {Springer Verlag},
address = {Porto, Portugal},
series = {LNCS},
abstract = {We propose mname, a method that reduces the number of required interactions between programs and tests in genetic programming. mname performs factorization of the matrix of the outcomes of interactions between the programs in a working population and the tests. Crucially, that factorization is applied to matrix that is only partially filled with interaction outcomes, i.e., sparse. The reconstructed approximate interaction matrix is then used to calculate the fitness of programs. In empirical comparison to several reference methods in categorical domains, mname attains higher success rate of synthesizing correct programs within a given computational budget.},
keywords = {}
}
We propose mname, a method that reduces the number of required interactions between programs and tests in genetic programming. mname performs factorization of the matrix of the outcomes of interactions between the programs in a working population and the tests. Crucially, that factorization is applied to matrix that is only partially filled with interaction outcomes, i.e., sparse. The reconstructed approximate interaction matrix is then used to calculate the fitness of programs. In empirical comparison to several reference methods in categorical domains, mname attains higher success rate of synthesizing correct programs within a given computational budget. |
Maciej Szkulmowski, Daniel Rumiński, Paweł Liskowski, Bartosz Wieloch, Krzysztof Krawiec, Bartosz Sikorski, Maciej D. Wojtkowski OCT Retinal Angiography Using Neural Networks (Inproceeding) The Annual Meeting of the Association for Research in Vision and Ophthalmology, 2016. (Links | BibTeX) @inproceedings{arvo:2016:oct,
title = {OCT Retinal Angiography Using Neural Networks},
author = {Maciej Szkulmowski and Daniel Rumiński and Paweł Liskowski and Bartosz Wieloch and Krzysztof Krawiec and Bartosz Sikorski and Maciej D. Wojtkowski},
url = {http://www.arvo.org/webs/am2016/sectionpdf/MOI/Session_130.pdf},
year = {2016},
date = {2016-01-01},
booktitle = {The Annual Meeting of the Association for Research in Vision and Ophthalmology},
keywords = {}
}
|
2015 |
Wojciech Jaśkowski, Paweł Liskowski, Marcin Szubert, Krzysztof Krawiec The performance profile: A multi-criteria performance evaluation method for test-based problems (Article) International Journal of Applied Mathematics and Computer Science, 26 (1), pp. 215–229, 2015. (Links | BibTeX) @article{jaskowski2016performance,
title = {The performance profile: A multi-criteria performance evaluation method for test-based problems},
author = {Wojciech Jaśkowski, Paweł Liskowski, Marcin Szubert and Krzysztof Krawiec},
url = {https://www.amcs.uz.zgora.pl/?action=paper&paper=886},
year = {2015},
date = {2015-01-01},
journal = {International Journal of Applied Mathematics and Computer Science},
volume = {26},
number = {1},
pages = {215--229},
keywords = {}
}
|
Krzysztof Krawiec, Paweł Liskowski Automatic Derivation of Search Objectives for Test-Based Genetic Programming (Inproceeding) Machado, Penousal; Heywood, Malcolm; McDermott, James; Castelli, Mauro; Garcia-Sanchez, Pablo; Burelli, Paolo; Risi, Sebastian; Sim, Kevin (Ed.): 18th European Conference on Genetic Programming, pp. 53–65, Springer, Copenhagen, 2015. (Abstract | Links | BibTeX) @inproceedings{Krawiec:2015:EuroGP,
title = {Automatic Derivation of Search Objectives for Test-Based Genetic Programming},
author = {Krzysztof Krawiec and Paweł Liskowski},
editor = {Penousal Machado and Malcolm I. Heywood and James McDermott and Mauro Castelli and Pablo Garcia-Sanchez and Paolo Burelli and Sebastian Risi and Kevin Sim},
url = {https://link.springer.com/chapter/10.1007/978-3-319-16501-1_5},
year = {2015},
date = {2015-01-01},
booktitle = {18th European Conference on Genetic Programming},
volume = {9025},
pages = {53--65},
publisher = {Springer},
address = {Copenhagen},
series = {LNCS},
abstract = {\"In genetic programming (GP)},
keywords = {}
}
"In genetic programming (GP) |
Marcin Szubert, Wojciech Jaśkowski, Paweł Liskowski, Krzysztof Krawiec The Role of Behavioral Diversity and Difficulty of Opponents in Coevolving Game-Playing Agents (Inproceeding) Mora, Antonio; Squillero, Giovanni (Ed.): Applications of Evolutionary Computation, pp. 394-405, Springer International Publishing, 2015, ISBN: 978-3-319-16548-6. (Abstract | Links | BibTeX) @inproceedings{SzubertEvoGames2015,
title = {The Role of Behavioral Diversity and Difficulty of Opponents in Coevolving Game-Playing Agents},
author = {Marcin Szubert, Wojciech Jaśkowski, Paweł Liskowski and Krzysztof Krawiec},
editor = {Mora, Antonio M. and Squillero, Giovanni},
url = {http://dx.doi.org/10.1007/978-3-319-16549-3_32},
isbn = {978-3-319-16548-6},
year = {2015},
date = {2015-01-01},
booktitle = {Applications of Evolutionary Computation},
volume = {9028},
pages = {394-405},
publisher = {Springer International Publishing},
series = {Lecture Notes in Computer Science},
abstract = {Generalization performance of learning agents depends on the training experience to which they have been exposed. In game-playing domains, that experience is determined by the opponents faced during learning. This analytical study investigates two characteristics of opponents in competitive coevolutionary learning: behavioral diversity and difficulty (performance against other players). To assess diversity, we propose a generic intra-game behavioral distance measure, that could be adopted to other sequential decision problems. We monitor both characteristics in two-population coevolutionary learning of Othello strategies, attempting to explain their relationship with the generalization performance achieved by the evolved solutions. The main observation is the existence of a non-obvious trade-off between difficulty and diversity, with the latter being essential for obtaining high generalization performance.},
keywords = {}
}
Generalization performance of learning agents depends on the training experience to which they have been exposed. In game-playing domains, that experience is determined by the opponents faced during learning. This analytical study investigates two characteristics of opponents in competitive coevolutionary learning: behavioral diversity and difficulty (performance against other players). To assess diversity, we propose a generic intra-game behavioral distance measure, that could be adopted to other sequential decision problems. We monitor both characteristics in two-population coevolutionary learning of Othello strategies, attempting to explain their relationship with the generalization performance achieved by the evolved solutions. The main observation is the existence of a non-obvious trade-off between difficulty and diversity, with the latter being essential for obtaining high generalization performance. |
Wojciech Jaśkowski, Marcin Szubert, Paweł Liskowski, Krzysztof Krawiec High-Dimensional Function Approximation for Knowledge-Free Reinforcement Learning: a Case Study in SZ-Tetris (Inproceeding) Proceedings of the 2015 Annual Conference on Genetic and Evolutionary Computation, pp. 567–573, ACM, Madrid, Spain, 2015, ISBN: 978-1-4503-3472-3. (Abstract | Links | BibTeX) @inproceedings{Jaskowski2015sztetris,
title = {High-Dimensional Function Approximation for Knowledge-Free Reinforcement Learning: a Case Study in SZ-Tetris},
author = {Wojciech Jaśkowski and Marcin Szubert and Paweł Liskowski and Krzysztof Krawiec},
url = {http://doi.acm.org/10.1145/2739480.2754783},
isbn = {978-1-4503-3472-3},
year = {2015},
date = {2015-01-01},
booktitle = {Proceedings of the 2015 Annual Conference on Genetic and Evolutionary Computation},
pages = {567--573},
publisher = {ACM},
address = {Madrid, Spain},
series = {GECCO '15},
abstract = {SZ-Tetris, a restricted version ofTetris, is a difficult reinforcement learning task. Previous researchshowed that, similarly to the original Tetris, value function-basedmethods such as temporal difference learning, do not work well forSZ-Tetris. The best performance in this game was achieved by employingdirect policy search techniques, in particular the cross-entropymethod in combination with handcrafted features. Nonetheless, a simpleheuristic hand-coded player scores even higher. Here we show that itis possible to equal its performance with CMA-ES (Covariance MatrixAdaptation Evolution Strategy). We demonstrate that furtherimprovement is possible by employing systematic n-tuple network, aknowledge-free function approximator, and VD-CMA-ES, a linear variantof CMA-ES for high dimension optimization. Last but not least, we showthat a large systematic n-tuple network (involving more than 4 millionparameters) allows the classical temporal difference learningalgorithm to obtain similar average performance to VD-CMA-ES, but at20 times lower computational expense, leading to the best policy forSZ-Tetris known to date. These results enrich the currentunderstanding of difficulty of SZ-Tetris, and shed new light on thecapabilities of particular search paradigms when applied torepresentations of various characteristics and dimensionality.},
keywords = {}
}
SZ-Tetris, a restricted version ofTetris, is a difficult reinforcement learning task. Previous researchshowed that, similarly to the original Tetris, value function-basedmethods such as temporal difference learning, do not work well forSZ-Tetris. The best performance in this game was achieved by employingdirect policy search techniques, in particular the cross-entropymethod in combination with handcrafted features. Nonetheless, a simpleheuristic hand-coded player scores even higher. Here we show that itis possible to equal its performance with CMA-ES (Covariance MatrixAdaptation Evolution Strategy). We demonstrate that furtherimprovement is possible by employing systematic n-tuple network, aknowledge-free function approximator, and VD-CMA-ES, a linear variantof CMA-ES for high dimension optimization. Last but not least, we showthat a large systematic n-tuple network (involving more than 4 millionparameters) allows the classical temporal difference learningalgorithm to obtain similar average performance to VD-CMA-ES, but at20 times lower computational expense, leading to the best policy forSZ-Tetris known to date. These results enrich the currentunderstanding of difficulty of SZ-Tetris, and shed new light on thecapabilities of particular search paradigms when applied torepresentations of various characteristics and dimensionality. |
Paweł Liskowski, Krzysztof Krawiec, Thomas Helmuth, Lee Spector Comparison of Semantic-aware Selection Methods in Genetic Programming (Inproceeding) Proceedings of the Companion Publication of the 2015 on Genetic and Evolutionary Computation Conference, pp. 1301–1307, ACM, Madrid, Spain, 2015, ISBN: 978-1-4503-3488-4. (Abstract | Links | BibTeX) @inproceedings{Liskowski:GECCO:2015,
title = {Comparison of Semantic-aware Selection Methods in Genetic Programming},
author = {Paweł Liskowski, Krzysztof Krawiec, Thomas Helmuth and Lee Spector},
url = {http://doi.acm.org/10.1145/2739482.2768505},
isbn = {978-1-4503-3488-4},
year = {2015},
date = {2015-01-01},
booktitle = {Proceedings of the Companion Publication of the 2015 on Genetic and Evolutionary Computation Conference},
pages = {1301--1307},
publisher = {ACM},
address = {Madrid, Spain},
series = {GECCO Companion '15},
abstract = {This study investigates the performance of several semantic-aware selection methods for genetic programming (GP). In particular, we consider methods that do not rely on complete GP semantics (i.e., a tuple of outputs produced by a program for fitness cases (tests)), but on binary outcome vectors that only state whether a given test has been passed by a program or not. This allows us to relate to test-based problems commonly considered in the domain of coevolutionary algorithms and, in prospect, to address a wider range of practical problems, in particular the problems where desired program output is unknown (e.g., evolving GP controllers). The selection methods considered in the paper include implicit fitness sharing (ifs), discovery of derived objectives (doc), lexicase selection (lex), as well as a hybrid of the latter two. These techniques, together with a few variants, are experimentally compared to each other and to conventional GP on a battery of discrete benchmark problems. The outcomes indicate superior performance of lex and ifs, with some variants of doc showing certain potential.},
keywords = {}
}
This study investigates the performance of several semantic-aware selection methods for genetic programming (GP). In particular, we consider methods that do not rely on complete GP semantics (i.e., a tuple of outputs produced by a program for fitness cases (tests)), but on binary outcome vectors that only state whether a given test has been passed by a program or not. This allows us to relate to test-based problems commonly considered in the domain of coevolutionary algorithms and, in prospect, to address a wider range of practical problems, in particular the problems where desired program output is unknown (e.g., evolving GP controllers). The selection methods considered in the paper include implicit fitness sharing (ifs), discovery of derived objectives (doc), lexicase selection (lex), as well as a hybrid of the latter two. These techniques, together with a few variants, are experimentally compared to each other and to conventional GP on a battery of discrete benchmark problems. The outcomes indicate superior performance of lex and ifs, with some variants of doc showing certain potential. |
2014 |
Wojciech Jaśkowski, Marcin Szubert, Paweł Liskowski Multi-Criteria Comparison of Coevolution and Temporal Difference Learning on Othello (Inproceeding) Esparcia-Alcazar,; Mora, (Ed.): EvoApplications 2014, pp. 301–312, Springer, 2014. (Abstract | Links | BibTeX) @inproceedings{Jaskowski2014multicriteria,
title = {Multi-Criteria Comparison of Coevolution and Temporal Difference Learning on Othello},
author = {Wojciech Jaśkowski and Marcin Szubert and Paweł Liskowski},
editor = {A. I. Esparcia-Alcazar and A. M. Mora},
url = {http://www.cs.put.poznan.pl/mszubert/pub/jaskowski2014evogames.pdf},
year = {2014},
date = {2014-01-01},
booktitle = {EvoApplications 2014},
volume = {8602},
pages = {301--312},
publisher = {Springer},
series = {Lecture Notes in Computer Science},
abstract = {We compare Temporal Difference Learning (TDL) with Coevolutionary Learning (CEL) on Othello. Apart from using three popular single-criteria performance measures: i) generalization performance or expected utility, ii) average results against a hand-crafted heuristic and iii) result in a head to head match, we compare the algorithms using performance profiles. This multi-criteria performance measure characterizes player’s performance in the context of opponents of various strength. The multi-criteria analysis reveals that although the generalization performance of players produced by the two algorithms is similar, TDL is much better at playing against the strong opponents, while CEL copes better against the weak ones. We also find out that TDL produces less diverse strategies than CEL. Our results confirm the usefulness of performance profiles as a tool for comparison of learning algorithms for games.},
keywords = {}
}
We compare Temporal Difference Learning (TDL) with Coevolutionary Learning (CEL) on Othello. Apart from using three popular single-criteria performance measures: i) generalization performance or expected utility, ii) average results against a hand-crafted heuristic and iii) result in a head to head match, we compare the algorithms using performance profiles. This multi-criteria performance measure characterizes player’s performance in the context of opponents of various strength. The multi-criteria analysis reveals that although the generalization performance of players produced by the two algorithms is similar, TDL is much better at playing against the strong opponents, while CEL copes better against the weak ones. We also find out that TDL produces less diverse strategies than CEL. Our results confirm the usefulness of performance profiles as a tool for comparison of learning algorithms for games. |
Paweł Liskowski, Krzysztof Krawiec Discovery of implicit objectives by compression of interaction matrix in test-based problems (Incollection) Parallel Problem Solving from Nature--PPSN XIII, pp. 611–620, Springer International Publishing, 2014. (Links | BibTeX) @incollection{liskowski2014discovery,
title = {Discovery of implicit objectives by compression of interaction matrix in test-based problems},
author = {Paweł Liskowski and Krzysztof Krawiec},
url = {https://link.springer.com/chapter/10.1007/978-3-319-10762-2_60},
year = {2014},
date = {2014-01-01},
booktitle = {Parallel Problem Solving from Nature--PPSN XIII},
pages = {611--620},
publisher = {Springer International Publishing},
keywords = {}
}
|
2013 |
Wojciech Jaśkowski, Paweł Liskowski, Marcin Szubert, Krzysztof Krawiec Improving coevolution by random sampling (Inproceeding) Proceedings of the 15th annual conference on Genetic and evolutionary computation, pp. 1141–1148, ACM 2013. (Links | BibTeX) @inproceedings{jaskowski2013improving,
title = {Improving coevolution by random sampling},
author = {Wojciech Jaśkowski and Paweł Liskowski and Marcin Szubert and Krzysztof Krawiec},
url = {https://dl.acm.org/citation.cfm?id=2463512},
year = {2013},
date = {2013-01-01},
booktitle = {Proceedings of the 15th annual conference on Genetic and evolutionary computation},
pages = {1141--1148},
organization = {ACM},
keywords = {}
}
|
Paweł Liskowski Quantitative analysis of the hall of fame coevolutionary archives (Inproceeding) Proceedings of the 15th annual conference companion on Genetic and evolutionary computation, pp. 1683–1686, ACM 2013. (Links | BibTeX) @inproceedings{liskowski2013quantitative,
title = {Quantitative analysis of the hall of fame coevolutionary archives},
author = {Paweł Liskowski},
url = {http://delivery.acm.org/10.1145/2490000/2482752/p1683-liskowski.pdf?ip=150.254.130.46&id=2482752&acc=ACTIVE%20SERVICE&key=6AF5E6E07E3D4A13%2EBC7592773EDED7D2%2E4D4702B0C3E38B35%2E4D4702B0C3E38B35&CFID=828947710&CFTOKEN=60818911&__acm__=1510649139_2e7d0acc25f4f01d66c5af791c9ddbb3},
year = {2013},
date = {2013-01-01},
booktitle = {Proceedings of the 15th annual conference companion on Genetic and evolutionary computation},
pages = {1683--1686},
organization = {ACM},
keywords = {}
}
|
Marcin Szubert, Wojciech Jaśkowski, Paweł Liskowski, Krzysztof Krawiec Shaping fitness function for evolutionary learning of game strategies (Inproceeding) Proceedings of the 15th annual conference on Genetic and evolutionary computation, pp. 1149–1156, ACM 2013. (Links | BibTeX) @inproceedings{szubert2013shaping,
title = {Shaping fitness function for evolutionary learning of game strategies},
author = {Marcin Szubert and Wojciech Jaśkowski and Paweł Liskowski and Krzysztof Krawiec},
url = {https://dl.acm.org/citation.cfm?id=2463513},
year = {2013},
date = {2013-01-01},
booktitle = {Proceedings of the 15th annual conference on Genetic and evolutionary computation},
pages = {1149--1156},
organization = {ACM},
keywords = {}
}
|
2012 |
Paweł Liskowski Co-evolution versus Evolution with Random Sampling for Acquiring Othello Position Evaluation (Mastersthesis) 2012. (BibTeX) @masterthesis{liskowski2012co,
title = {Co-evolution versus Evolution with Random Sampling for Acquiring Othello Position Evaluation},
author = {Paweł Liskowski},
year = {2012},
date = {2012-01-01},
address = {Poznań, Poland},
school = {Poznan University of Technology},
keywords = {}
}
|