J. Šíma, P. Mančal, J. Bouzek: Numerical Processing of
Results of Quantitative Immunochemical Analytical Methods on Microcomputer. Prague: ÚSOL publisher,
1985. (in Czech)
J. Šíma: Generalized back propagation for training pattern
derivatives. Neural Network World 4(1):91-98, 1994.
J. Šíma, R. Neruda: Neural networks as expert systems. Neural
Network World 2(6):775-784, 1992.
J. Šíma: Generalized back propagation for interval training
patterns. Neural Network World 2(2):167-174, 1992.
J. Šíma: The multi-layered neural network as an adaptive expert
system with the ability to work with incomplete information and
to provide justification of inference. Neural Network World
2(1):47-58, 1992.
J. Šíma, P. Vidnerová: Weight-rounding error in deep neural networks. Proceedings of the ECML PKDD 2025 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Part IV, LNCS 16016, pp. 398-416, Springer, 2025. DOIslides (photo) poster (photo)
J. Šíma, J. Cabessa: The power of max pooling layer. Proceedings of the ICANN 2025 Thirty-Fourth International Conference on Artificial Neural Networks and Machine Learning, Part I, LNCS 16068, pp. 221-233, Springer, 2025. DOIslides
J. Šíma, J. Cabessa: Energy complexity of fully-connected layers. Proceedings of the IWANN 2023 Seventeenth International Work-Conference on Artificial Neural Networks, Part I, LNCS 14134, pp. 3-15, Springer, 2023. DOIslides
J. Šíma, P. Vidnerová, V. Mrázek: Energy complexity model for convolutional neural networks. Proceedings of the ICANN 2023 Thirty-Second International Conference on Artificial Neural Networks and Machine Learning, Part X, LNCS 14263, pp. 186-198, Springer, 2023. DOIslides
M. Plátek, F. Mráz, D. Pardubská, D. Průša, J. Šíma:
On separations of LR(0)-grammars by
two types of pumping patterns. Proceedings of the ITAT 2021 Twenty-First Conference on Information Technologies
- Applications and Theory, Workshop on Automata, Formal and Natural Languages, CEUR Workshop Proceedings, Vol.
2962, pp. 140-146, 2021.
F. Mráz, D. Pardubská, M. Plátek, J. Šíma:
Pumping deterministic monotone restarting automata and DCFL. Proceedings of the ITAT 2020 Twentieth Conference
on Information Technologies - Applications and Theory, Workshop on Automata, Formal and Natural Languages, CEUR
Workshop Proceedings, Vol. 2718, pp. 51-58, 2020.
J. Šíma:
Counting with analog neurons. Proceedings of the ICANN 2019 Twenty-Eighth
International Conference on Artificial Neural Networks, LNCS 11727, Part I, pp. 389-400, Springer, 2019. DOIposter
J. Šíma:
Three analog neurons are Turing universal. Proceedings of the TPNC 2018 Seventh
International Conference on the Theory and Practice of Natural Computing, LNCS 11324, pp. 460-472, Springer,
2018. DOIslides
J. Šíma, P. Savický:
Cut languages in rational bases. Proceedings of the LATA 2017 Eleventh
International Conference on Language and Automata Theory and Applications, LNCS 10168, pp. 311-322, Springer,
2017. DOIslides
J. Šíma:
The power of extra analog neuron. Proceedings of the TPNC 2014 Third International
Conference on the Theory and Practice of Natural Computing, LNCS 8890, pp. 243-254, Springer, 2014. DOIslides
S. Žák, J. Šíma:
A Turing machine distance hierarchy. Proceedings of the LATA 2013 Seventh
International Conference on Language and Automata Theory and Applications, LNCS 7810, pp. 570-578, Springer,
2013. DOIslides
J. Šíma:
Gradient learning in networks of smoothly spiking neurons. Proceedings of the
ICONIP 2008 Fifteenth International Conference on Neural Information Processing of the Asia-Pacific Neural
Network Assembly, LNCS 5507, Part II, pp. 179-186, Springer, 2009. DOI
J. Šíma:
Gradient learning in networks of smoothly spiking neurons. (extended abstract) Proceedings of the ICONIP 2008 Fifteenth International Conference on Neural Information Processing of the
Asia-Pacific Neural Network Assembly, pp. 167-168, The Knowledge Engineering and Discovery Research Institute
(KEDRI), Auckland University of Technology, 2008.
J. Šíma:
Computational resources in neural network models. A tutorial at the ICONIP 2008 Fifteenth International Conference on Neural Information Processing of the Asia-Pacific Neural Network Assembly, November, Auckland, New Zealand, 2008.
J. Šíma, S.E. Schaeffer:
On the NP-completeness of some graph cluster measures.
Proceedings of the SOFSEM 2006 Thirty-Second International Conference on Current Trends in Theory and Practice
of Informatics, LNCS 3831, pp. 530-537, Springer, 2006. DOI
J. Šíma: Review of integration strategies in neural hybrid systems. Euro-International Symposium on
Computational Intelligence (E-ISCI'2000), In P. Sinčák, V. Vaščák (eds.): Quo Vadis Computational Intelligence?
New Trends and Approaches in Computational Intelligence, Studies in Fuzziness and
Soft Computing, Vol. 54, pp. 355-360, Springer, 2000.
J. Šíma, P. Orponen: A continuous-time Hopfield net simulation of discrete neural networks. Proceedings of the
NC'2000 Second International ICSC Symposium on Neural Computation, pp. 36-42, Wetaskiwin (Canada): ICSC Academic
Press, 2000.
J. Šíma, P. Orponen, T. Antti-Poika: Some afterthoughts on Hopfield networks. Proceedings of the
SOFSEM 1999 Seminar on Current Trends in Theory and Practice of Informatics, LNCS 1725, pp. 459-469, Springer,
1999. DOI
J. Šíma: Complexity issues in neurocomputing. Invited talk at the International Students' Workshop on Neural
Networks, Faculty of Mathematics and Physics, Charles University, Prague, 1998.
J. Šíma: Aunt's problem: Table rounding. Proceedings of the SOFSEM 1996 Seminar on Current Trends in Theory and
Practice of Informatics, LNCS 1175, pp. 441-450, Springer, 1996. DOI
J. Šíma, J. Wiedermann: Neural Language Acceptors. Proceedings of the DLT 1995 Second International Conference
Developments in Language Theory, pp. 430-439, Singapore: World Scientific, 1996.
J. Šíma: Hopfield languages. Proceedings of the SOFSEM Seminar on Current Trends in Theory and Practice of
Informatics, LNCS 1012, pp. 461-468, Springer, 1995. DOI
J. Šíma, R. Neruda: The empty expert system and its application in medicine. Proceedings of the Twelfth European
Meeting on Cybernetics and Systems Research, pp. 1825-1832, Singapore: World Scientific, 1994.
J. Šíma: Neural learning is not efficient. Proceedings of the SOFSEM 1994 winter school, pp. 107-110, 1994.
J. Wiedermann, J. Šíma, R. Neruda: Massaging heapsort into an optimal disk sorting algorithm. Proceedings of the
SOFSEM 1994 winter
school, pp. 117-120, 1994.
J. Šíma, R. Neruda: EXPSYS - a tool for neural expert system design. Proceedings of the NEURONET conference,
Prague, 1993.
J. Šíma, R. Neruda: Neural Expert Systems. Proceedings of the IJCNN'92 International Joint Conference on Neural
Networks, Vol. 3, pp. 634-639, Publishing House of Electronics Industry, 1992.
2010-2014 NoSCoM: Non-Standard Computational Models
and Their Applications in Complexity, Linguistics, and Learning, Czech
Science Foundation (GA ČR), GAP202/10/1333, principal investigator
2005-2009 Methods for Intelligent Systems
and Their Applications in Datamining and Natural Language
Processing, Czech Academy of Sciences, Program "Information Society", 1ET100300517,
principal investigator, final evaluation: the scientific aim
fulfilled with outstanding results
2000-2002 NEUROINFORMATICS: Computational Theory of Neural Networks, Czech Academy of Sciences, IAB2030007, principal investigator, final evaluation: the
scientific aim fulfilled with outstanding results
2003-2006: INTAS Council of Scientists (The International
Association for the Promotion of Cooperation with Scientists from the
New Independent States of the Former Soviet Union), Brussels
