prof. dr hab. Jacek Tabor dr hab. Igor T. Podolak dr Jarek Duda dr Krzysztof Misztal dr hab. Adam Roman dr Marek Śmieja dr Przemysław Spurek dr Marcin Żelawski dr Bartosz Zieliński dr Wojciech M. Czarnecki
mgr Stanisław Jastrzębski mgr Damian Leśniak mgr Maciej J. Mikulski mgr Agnieszka Pocha mgr Igor Sieradzki mgr Magdalena Wiercioch mgr Konrad Żołna mgr Jakub Chłędowski mgr Piotr Kowenzowski mgr Łukasz Maziarka


Igor Podolak, Adam Roman, Marek Szykuła, Bartosz Zieliński A machine learning approach to synchronization of automata, EXPERT SYST APPL (2018), 357-371
A Roman, K Zmitrowicz, Inżynieria wymagań: studium przypadkw, Warszawa: Wydawnictwo Naukowe PWN (2018)
S Schmitt, JJ Hudson, A Zidek, S Osindero, C Doersch, WM Czarnecki, ..., Kickstarting Deep Reinforcement Learning, arXiv preprint arXiv:1803.03835 (2018)


Andrzej Bojarski, Wojciech Czarnecki, Stanisław Jastrzębski, Damian Leśniak, Sabina Podlewska Quo vadis G Protein-Coupled Receptor ligands? A tool for analysis of the emergence of new groups of compounds over time, BIOORG MED CHEM LETT vol. 27 (2017), 626-631
YW Teh, V Bapst, WM Czarnecki, J Quan, J Kirkpatrick, R Hadsell, ..., Distral: Robust Multitask Reinforcement Learning, NIPS 2017 (2017)
WM Czarnecki, G Świrszcz, M Jaderberg, S Osindero, O Vinyals, ..., Understanding Synthetic Gradients and Decoupled Neural Interfaces, ICML 2017 (2017)
WM Czarnecki, S Osindero, M Jaderberg, G Świrszcz, R Pascanu, Sobolev Training for Neural Networks, NIPS 2017 (2017)
M Jaderberg, WM Czarnecki, S Osindero, O Vinyals, A Graves, ..., Decoupled neural interfaces using synthetic gradients, ICML 2017 (2017)
M Jaderberg, V Mnih, WM Czarnecki, T Schaul, JZ Leibo, D Silver, ..., Reinforcement learning with unsupervised auxiliary tasks, ICLR 2017 (2017)
Marek Śmieja, Magdalena Wiercioch Constrained clustering with a complex cluster structure, ADV DATA ANAL CLASSI vol. 11/3 (2017), 493-518
A Roman, K Zmitrowicz, Testowanie oprogramowania w praktyce: studium przypadkw, Warszawa: Wydawnictwo Naukowe PWN (2017)
Adam Roman Data Validation Using Model-Based Testing and Finite Automata Synchronization, AIP CONFERENCE PROCEEDINGS 1836 (2017), 020007
Wojciech Czarnecki, Stanisław Jastrzębski, Damian Leśniak How to evaluate word embeddings? On importance of data efficiency and simple supervised tasks, ARXIV (2017),
Jakub Kowalski, Adam Roman A New Evolutionary Algorithm for Synchronization vol. LNCS 10199 (2017), "Applications of Evolutionary Computation", Springer Verlag (połaczony z Kluwer Academic Publishing)
Adam Roman Inżynieria wymagań w praktyce. Studium przypadków, (2017), PWN
Magdalena Wiercioch Towards Learning Word Representation, SCHEDAE INFORMATICAE vol. 25 (2017), 103–115
M Jaderberg, V Dalibard, S Osindero, WM Czarnecki, J Donahue, ..., Population Based Training of Neural Networks, arXiv preprint arXiv:1711.09846 (2017)
Magdalena Wiercioch Feature Selection in Texts, PROCEEDINGS OF THE 10TH INTERNATIONAL CONFERENCE ON COMPUTER RECOGNITION SYSTEMS CORES 2017 vol. 578. Advances in Intelligent Systems and Computing (2017), 336-345
K Janocha, WM Czarnecki, On loss functions for deep neural networks in classification, arXiv preprint arXiv:1702.05659 (2017)
Adam Roman Testowanie w praktyce. Studium przypadków 2.0., (2017), PWN
P Sunehag, G Lever, A Gruslys, WM Czarnecki, V Zambaldi, M Jaderberg, ..., Value-Decomposition Networks For Cooperative Multi-Agent Learning, arXiv preprint arXiv:1706.05296 (2017)
S Podlewska, WM Czarnecki, R Kafel, AJ Bojarski, Creating the new from the old: Combinatorial libraries generation with machine-learning-based compound structure optimization, Journal of chemical information and modeling 57 (2), 133-147 (2017)


Wojciech Czarnecki, Stanisław Jastrzębski, Damian Leśniak Learning to SMILE(S), INTERNATIONAL CONFERENCE ON LEARNING REPRESENTATIONS (WORKSHOP TRACK) (2016),
Adam Roman, Vojtech Vorel Complexity of road coloring with prescribed reset words, J COMPUT SYST SCI (2016), 30
Maciej Chmielarz, Adam Roman, Lucjan Stapp Postgraduate Studies on Software Testing in Poland, SOFTWARE ENGINEERING: IMPROVING PRACTICE THROUGH RESEARCH (2016), 177-186
Rafał Babiarz, Adam Roman Dynamic Stylometry for Defect Prediction, ADVANCES IN INTELLIGENT SYSTEMS AND COMPUTING (2016), 91-102
Wojciech Czarnecki, Jacek Tabor Online Extreme Entropy Machines for Streams Classification and Active Learning, ADVANCES IN INTELLIGENT SYSTEMS AND COMPUTING vol. 403 (2016), 371-381
Michał Mnich, Adam Roman, Piotr Wawrzyniak Mutation churn model, SCHEDAE INFORMATICAE vol. 25 (2016), 227-236
Adam Roman, Karolina Zmitrowicz Testowanie w praktyce. Studium przypadków (redakcja), (2016), PWN
G Swirszcz, WM Czarnecki, R Pascanu, Local minima in training of neural networks, arXiv preprint arXiv:1611.06310 (2016)


Wojciech Czarnecki, Jacek Tabor Multithreshold Entropy Linear Classifier: Theory and Applications, EXPERT SYST APPL vol. 42 (2015), 5591-5606
Andrzej Bojarski, Wojciech Czarnecki, Sabina Smusz, Dawid Warszycki Exploiting uncertainty measures in compounds activity prediction using support vector machines, BIOORG MED CHEM LETT vol. 25 (2015), 100-105
Adam Roman, Marek Szykuła Forward and backward synchronizing algorithms, EXPERT SYST APPL vol. 42(24) (2015), 9512-9527
Andrzej Bojarski, Wojciech Czarnecki, Sabina Podlewska Robust optimization of SVM hyperparameters in the classification of bioactive compounds, J CHEMINFORMATICS vol. 7 (38) (2015), 15
Wojciech Czarnecki, Jacek Tabor Extreme entropy machines: robust information theoretic classification, PATTERN ANAL APPL (2015), 1-18
Wojciech Czarnecki Weighted Tanimoto Extreme Learning Machine with Case Study in Drug Discovery, IEEE COMPUT INTELL M vol. 10 (3) (2015), 19-29
Adam Roman, Vojtech Vorel Parameterized Complexity of Synchronization and Road Coloring, DISCRETE MATH THEOR vol. 17(1) (2015), 307-330
W Czarnecki, Renyi's quadratic entropy in classification, (2015)
Wojciech Czarnecki Adaptive Active Learning with Ensemble of Learners and Multiclass Problems, LECTURE NOTES IN COMPUTER SCIENCE vol. 9119 (2015), 415 - 426
Marek Śmieja, Magdalena Wiercioch Mixture of metrics optimization for machine learning problems, SCHEDAE INFORMATICAE vol. 24 (2015), 133-142
Marek Śmieja, Jacek Tabor, Magdalena Wiercioch Probability Index of Metric Correspondence as a measure of visualization reliability, PROCEEDINGS OF ECML PKDD WORKSHOP ON MACHINE LEARNING IN LIFE SCIENCES (2015), 16-27
Wojciech Czarnecki, Stanisław Jastrzębski, Sabina Podlewska, Igor Sieradzki Active Learning of Compounds Activity - Towards Scientifically Sound Simulation of Drug Candidates Identification, PROCEEDINGS OF 2ND WORKSHOP ON MACHINE LEARNING IN LIFE SCIENCES (2015), 40-51
Wojciech Czarnecki, Stanisław Jastrzębski Analysis of compounds activity concept learned by SVM using robust Jaccard based low-dimensional embedding, SCHEDAE INFORMATICAE vol. 24 (2015), 9-19
Wojciech Czarnecki, Rafał Józefowicz, Jacek Tabor Maximum Entropy Linear Manifold for Learning Discriminative Low-dimensional Representation, LECTURE NOTES IN ARTIFICIAL INTELLIGENCE vol. 9284 (2015), 52-67
Wojciech Czarnecki, Krzysztof Rataj Compounds Activity Prediction in Large Imbalanced Datasets with Substructural Relations Fingerprint and EEM, IEEE TRUSTCOM/BIGDATASE/ISPA PROCEEDINGS vol. 2 (2015), 192-197
Wojciech Czarnecki, Rafał Józefowicz Fast optimization of Multithreshold Entropy Linear Classifier, SCHEDAE INFORMATICAE vol. 23 (2015), 57–67
Adam Roman, Vojtech Vorel Complexity of Road Coloring with Prescribed Reset Words, LECTURE NOTES IN COMPUTER SCIENCE vol. 8977 (2015), 161-172
Andrzej Bojarski, Wojciech Czarnecki, Sabina Podlewska Extremely Randomized Machine Learning Methods for Compound Activity Prediction, MOLECULES vol. 20 (11) (2015), 20107-20117
Adam Roman Programista jako tester, COMPUTERWORLD (2015), 24-25
Wojciech Czarnecki On the Consistency of Multithreshold Entropy Linear Classifier, SCHEDAE INFORMATICAE vol. 24 (2015), 123-132
Wojciech Czarnecki, Rafał Józefowicz Neural Connectivity Reconstruction from Calcium Imaging Signal using Random Forest with Topological Features, JMLR: WORKSHOP AND CONFERENCE PROCEEDINGS vol. 46 (2015), 67-76
Adam Roman Testowanie i jakość oprogramowania. Modele, techniki, narzędzia, (2015), PWN
Mateusz Drewienkowski, Adam Roman A complete solution to the complexity of Synchronizing Road Coloring for non-binary alphabets, INFORM COMPUT vol. 242 (2015), 383-393
Adam Roman Testowanie i analiza mutacyjna, PROGRAMISTA vol. 6(37) (2015), 52-55


Wojciech Czarnecki, Jacek Tabor Two ellipsoid Support Vector Machines, EXPERT SYST APPL vol. 41 (18) (2014), 8211-8224
Wojciech Czarnecki, Igor Podolak Adaptive Active Learning as a Multi-armed Bandit Problem, FRONTIERS IN ARTIFICIAL INTELLIGENCE AND APPLICATIONS vol. 263 (2014), 989-990
Wojciech Czarnecki, Krzysztof Szarzyński, Andrzej Wójtowicz Designing a competition for autonomous robots with a restricted set of sensors with case study of LEGO NXT, JOURNAL OF AUTOMATION, MOBILE ROBOTICS & INTELLIGENT SYSTEMS - JAMRIS (2014), 76-81
Agata Dróżdż, Marzena Frołow, Agata Kowalewska, Adam Roman, Bartosz Zieliński A new approach to automatic continuous artery diameter measurement , PROCEEDINGS OF THE 2014 FEDERATED CONFERENCE ON COMPUTER SCIENCE AND INFORMATION SYSTEMS vol. 2 (2014), 247–251
S Smusz, W Czarnecki, D Warszycki, AJ Bojarski, Uncertainty of the in vitro experiments in the construction of predictive models, GPCR Spring Conference 2014 (2014)
WM Czarnecki, J Tabor, Cluster based RBF Kernel for Support Vector Machines, arXiv preprint arXiv:1408.2869 (2014)
WM Czarnecki, InFeST–ImageJ Plugin for Rapid Development of Image Segmentation Pipelines, Man-Machine Interactions 3, 283-290 (2014)
Adam Roman Testowanie eksploracyjne - cudowny lek czy zwykłe placebo?, SOFTWARE DEVELOPER'S JOURNAL (2014), 4-15


Igor Podolak, Adam Roman Theoretical foundations and practical results for a hierarchical classifier with overlapping clusters, COMPUT INTELL-US (2013), 357-388
Wojciech Czarnecki, Igor Podolak Machine Learning with Known Input Data Uncertainty Measure, LECTURE NOTES IN COMPUTER SCIENCE (2013), 379-388
WM Czarnecki, K Szarzyński, A Wójtowicz, ʥ S C S ʥ LEGO NXT, (2013)
IT Podolak, A Roman, Theoretical foundations and experimental results for a hierarchical classifier with overlapping clusters, Computational Intelligence 29 (2), 357-388 (2013)
Kamil Bartocha, William Beckler, Krzysztof Osiewalski, Adam Roman’s 12-week geo-experiment reveals a 43% uplift in Google AdWords values, GOOGLE ADWORDS BLOG, ADWORDS.BLOGSPOT.COM (2013), 3


Rafał Kudłacik, Adam Roman, Hubert Wagner Effective synchronizing algorithms, EXPERT SYST APPL vol. 39 (2012), 11746-11757
Dariusz Jędrzejczyk, Igor Podolak, Adam Roman Application of Hierarchical Classifier to Minimal Synchronizing Word problem, LECTURE NOTES IN COMPUTER SCIENCE vol. 7267 (2012), 421-429
Z Vetulani, E Geoffrois, W Czarnecki, B Kochanowski, Language Resources for Public Security Applications: Needs and Specificities, Language Resources and Evaluation Conference 2011 - Language Resources for… (2012)
Adam Roman O 'Epistemologii' Jana Woleńskiego - spojrzenie matematyka, DIAMETROS - AN ONLINE JOURNAL OF PHILOSOPHY (2012), 175-188
Adam Roman P--NP threshold for the Synchronizing Road Coloring, LECTURE NOTES IN COMPUTER SCIENCE vol. 7183 (2012), 482-492
A Roman, P–NP threshold for synchronizing road coloring, International Conference on Language and Automata Theory and Applications… (2012)
Wojciech Czarnecki Multilayer Neural Networks with Receptive Fields as a Model for the Neuron Reconstruction Problem, LECTURE NOTES IN COMPUTER SCIENCE (2012), 242-250


Igor Podolak, Adam Roman Fusion of supervised and unsupervised training methods for a multi-class classification problem, PATTERN ANAL APPL vol. 14(4) (2011), 395-413
IT Podolak, A Roman, CORES: fusion of supervised and unsupervised training methods for a multi-class classification problem, Pattern Analysis and Applications 14 (4), 395-413 (2011)
Igor Podolak, Adam Roman Risk function estimation for subproblems in a hierarchical classifier, PATTERN RECOGN LETT vol. 32(15) (2011), 2136-2142
A Roman, The NP-completeness of the road coloring problem, Information Processing Letters 111 (7), 342-347 (2011)
Igor Podolak, Adam Roman Risk estimation for hierarchical classifier, LECTURE NOTES IN ARTIFICIAL INTELLIGENCE vol. 6678 (2011), 156-163
Maria Foryś, Adam Roman Zbiór zadań z teorii języków formalnych i automatów, (2011), Wydawnictwo Uniwersytetu Jagiellońskiego
Adam Roman NP-completeness of the Road Coloring Problem, INFORM PROCESS LETT vol. 111 (2011), 342-347
Agnieszka Mielniczek (Deszyńska), Igor Podolak, Adam Roman On the number of clusterings in a hierarchical classification model with overlapping clusters, SCHEDAE INFORMATICAE vol. 20 (2011), 137-157
Krzysztof Chmiel, Adam Roman COMPAS - a computing package for synchronization, LECTURE NOTES IN COMPUTER SCIENCE vol. 6482 (2011), 79-86


Adam Roman Experiments on synchronizing automata, SCHEDAE INFORMATICAE vol. 19 (2010), 35-51


Adam Roman Genetic Algorithm for Synchronization, LECTURE NOTES IN COMPUTER SCIENCE vol. 5457 (2009), 684-695
A Roman, Synchronizing finite automata with short reset words, Applied Mathematics and Computation 209 (1), 125-136 (2009)
Igor Podolak, Adam Roman A new notion of weakness in classification theory vol. 57 (2009), "Advances in Intelligent and Soft Computing", Springer Verlag (połaczony z Kluwer Academic Publishing)
Adam Roman Synchronizing Finite Automaton with Short Reset Words, APPL MATH COMPUT vol. 209 (2009), 125-136
Adam Roman Decision Version of the Road Coloring Problem is NP-complete, LECTURE NOTES IN COMPUTER SCIENCE vol. 5699 (2009), 287-297


Wit Foryś, Adam Roman Lower bound for the length of synchronizing words in partially-synchronizing automata, LECTURE NOTES IN COMPUTER SCIENCE vol. 4910 (2008), 448 - 459
Bartosz Bierkowski, Piotr Kalita, Igor Podolak, Adam Roman Algorithm for intelligent prediction of requests in business systems, LECTURE NOTES IN COMPUTER SCIENCE vol. 4910 (2008), 696-707
A Roman, A note on Černy conjecture for automata over 3-letter alphabet, Journal of Automata, Languages and Combinatorics 13 (2), 141-143 (2008)


A Roman, Two Simple Methods for Obtaining New Classes of Automata Fulfilling ˇCern Conjecture, Schedae Informaticae 16 (2007)


A Roman, Merging states and synchronization problem, Schedae Informaticae 15, 95-108 (2006)


A Roman, New Algorithms for Finding Short Reset Sequences in Synchronizing Automata., IEC (Prague), 13-17 (2005)


A Roman, The proposal of Algorithmic Turing Test, (2004)

Contact Us

Igor Podolak, PhD
Group of Machine Learning Research
Faculty of Mathematics and Computer Science
ul. Lojasiewicza 6
30-342, Cracow, Poland