Statistical Analysis and Modeling Group

Welcome to the Statistical Analysis and Modeling Group! The Group is a part of the Institute of Computer Science of the Polish Academy of Sciences. The Group's research activities concern probabilistic and statistical modeling of natural phenomena and statistical inference for constructed models.

The Group maintains strong links with the Faculty of Mathematics and Information Sciences of the Warsaw University of Technology where several of its members teach courses and pursue joint research.

People Research Publications

Recent news

New publication:
Łukasz Dębowski, Universal Densities Exist for Every Finite Reference Measure, IEEE Transactions on Information Theory, 2023.

New publications:
Nikolay Bazhenov, Dariusz Kalociński, Degree Spectra, and Relative Acceptability of Notations, 31st EACSL Annual Conference on Computer Science Logic (CSL 2023), 2023.

Tae-Won Lee, Paweł Teisseyre, Jaesung Lee, Effective Exploitation of Macroeconomic Indicators for Stock Direction Classification using the Multimodal Fusion Transformer, IEEE Access, 2023.

Paweł Teisseyre, Jaesung Lee, Multilabel all-relevant feature selection using lower bounds of conditional mutual information, Expert Systems with Applications, 2023.

New publications:
Dariusz Kalociński and Michał Wrocławski, Generalization of Shapiro’s theorem to higher arities and noninjective notations, Archive for Mathematical Logic, 2022.

Uplift modeling workshop and tutorial Szymon Jaroszewicz is coorganizing an uplift modeling workshop and tutorial at the ECML/PKDD'22 conference. Submission deadline is June 30th.

New publications:
Dariusz Kalociński (co-authors: Nikolay Bazhenov, Michał Wrocławski), Intrinsic Complexity of Recursive Functions on Natural Numbers with Standard Order, 39th International Symposium on Theoretical Aspects of Computer Science (STACS 2022).
Ewelina Pośpiech, Paweł Teisseyre, Jan Mielniczuk, Wojciech Branicki, Predicting Physical Appearance from DNA Data—Towards Genomic Solutions, Genes, 2022.

New publications:
Łukasz Dębowski, A Refutation of Finite-State Language Models through Zipf’s Law for Factual Knowledge , Entropy, 2021.
Iván G. Torre, Łukasz Dębowski, Antoni Hernández-Fernández, Can Menzerath’s law be a criterion of complexity in communication?, PLOS ONE, 2021.
M. Łazęcka, J. Mielniczuk, P. Teisseyre, Estimating the class prior for positive and unlabelled data via logistic regression, Advances in Data Analysis and Classification, 2021.

New publication:
Mariusz Kubkowski, Jan Mielniczuk, Paweł Teisseyre, How to Gain on Power: Novel Conditional Independence Tests Based on Short Expansion of Conditional Mutual Information, Journal of Machine Learning Research, 2021

The hardcover edition of book "Information Theory Meets Power Laws: Stochastic Processes and Language Models" by Łukasz Dębowski will be sold in Europe starting from April 2021 (Amazon UK). It can be already purchased in the US (Wiley).

Szymon Jaroszewicz received the title of full professor.

New publications:
Barbara Żogała-Siudem, Szymon Jaroszewicz: Fast stepwise regression based on multidimensional indexes. Information Sciences 549 (2021)
Paweł Teisseyre: Classifier chains for positive unlabelled multi-label learning. Knowledge-Based Systems 213, (2021).

Krzysztof Rudaś was chosen as the Polish representative at the European Young Statisticians Meetings

Soon to be released monograph "Information Theory Meets Power Laws: Stochastic Processes and Language Models" by Łukasz Dębowski can be pre-ordered at Wiley (scheduled for February 2021)

Szymon Jaroszewicz has been invited to the IJCAI 2021 program committee as Senior Program Committee member (SPC).

New preprint by Łukasz Dębowski entitled 'On a Class of Markov Order Estimators Based on PPM and Other Universal Codes' is now avaialble at arXiv.

A new paper entitled 'Approximating Information Measures for Fields' by Łukasz Dębowski has been published in journal 'Entropy'. Paper can be found here

Szymon Jaroszewicz has been invited to the Program Committee of the ACM SIGKDD'20 conference.

Szymon Jaroszewicz has been elected to the Committee on Informatics of the Polish Academy of Sciences.

A new paper entitled 'Learning classifier chains using matrix regularization: application to multimorbidity prediction' by Paweł Teisseyre has been accepted for publication in ECAI2020 proceedings.

Next seminar February 24, 3pm: Łukasz Dębowski (ICS PAS), title: 'Losowość algorytmiczna: Podstawowe idee i problemy (cz. II)'.

Next seminar February 10, 3pm: Łukasz Dębowski (ICS PAS), title: 'Losowość algorytmiczna: Podstawowe idee i problemy (cz. I)'.

Next seminar January 13, 3pm: Piotr Przybyła (ICS PAS), title: 'Pretrenowane modele językowe od podstaw '.

A new paper entitled 'Shrinkage Estimators for Uplift Regression' by Krzysztof Rudaś and Szymon Jaroszewicz has been published in ECML2019 proceedings. Paper can be found here