Dr Jeremi Ochab

Contact

room D-2-10
jeremi.ochab[at]uj.edu.pl
Tel: +48 12 664 4797

Jeremi Ochab photo


Google Scholar
ResearchGate
Academia

Research interests

In general search for applications of statistical physics and machine learning to neurophysiological and behavioural data (EEG, fMRI, actigraphy, etc.) and to linguistic data (style and authorship of literary texts; see, project computationalstylistics.github.io). I have conducted research on: detection of modules in complex networks/graphs, random walks (http://arxiv.org/abs/1004.3667), synchronisation, and epidemic modelling. I am also tangentially interested in linguistics and translation.

My most recent CV is available here.

General information

I graduated in theoretical physics (as a part of Interdisciplinary Studies in Mathematics and Natural Sciences programme) and took my doctorate at the Jagiellonian University. At the moment I am an assistant professor at the Department of Theory of Complex Systems, collaborating with prof. Maciej A. Nowak and prof. Dante R. Chialvo, as well as a member of Computational stylistics research group at the Institute of Polish Language, Polish Academy of Sciences. I have been a principal investigator in grants financed by the National Science Centre of Poland: Sonata 9, titled Application of the complex systems theory in multidimensional fluctuation analysis of human brain EEG signals (grant no. 2015/17/D/ST2/03492) and Preludium 5, titled Statistical foundations in detection of modular structures in complex networks (grant no. 2013/09/N/ST6/01419), and I worked in in a Maestro grant Interdisciplinary applications of random matrix theory. My Phd studies (2009-2013) were funded from the scholarship of the Foundation for Polish Science (Fundacja Nauki Polskiej www.fnp.org.pl), as the project Dynamical systems on complex networks within the Jagellonian University International PhD Studies in Physics of Complex Systems (http://www.mpd.if.uj.edu.pl) under supervision of prof. Z. Burda. I have been awarded the scholarship for PhD students for outstanding achievements by the Polish Ministry of Science and Higher Education in 2013.

Teaching

This year I am (co-)teaching courses in:

I have conducted the courses:

  • AI workshop I (with K. Kutt, S. Bobek, G. Nalepa),
  • deep learning (in English),
  • logic and set theory (classes, 1st y. students of applied computer science, 1 semester),
  • computational methods in physics (classes, 1st y. students of physics, 1 semester),
  • introduction to data analysis (lab in English, 2nd-3rd y. students of applied computer science, 1 semester),
  • statistical field theory (classes, 4th y. students of theoretical physics, 1 semester),
  • discrete mathematics (classes, 1st/2nd y. students of applied computer science, 1 semester),
  • supplementary lectures in mathematics  (classes, 1st y.  students  of physics/biophysics/..., 1 semester),
  • 1st physics laboratory (1 semester).

Master and bachelor theses

Let's talk, if you are interested in:

Neuroimaging – neuroimaging – neuropsychology

N1. Prediction of erroneous responses from human EEG
N2. Correlatiing neurophysiological, actigraphic, and behavioural data in chronic sleep deficit
N3. Classification of human chronotypes based on genes, actigraphy, and neuroimaging
N4. Pose estimation models for sport analysis and data-driven sports training

Text-mining – linguistics - digital humanities

L0. Automatyczne speech recognition for large-scale phonetic transcription
L1. Classifier of literary style chronology
L2. Extracting interpretable/explainable features from literary style classifiers
L3. Diachronic word embeddings (embedding space alignment)
L2. Classifier of text authorship based on time-series analysis
L4. Text-mining i machine learning of papyri
L5. Complex network analysis for text clustering

Current students:

  • mgr M. Kądzioła, L5. Supporting recognition and authorial profiling of spoken texts with computational stylistics, linguistics (co-supervisor, IPL PAS)
  • mgr M. Lotka, Methods of inferring correlations from fMRI signals, physics

Previous students:

Recent papers

Theses

Translations

Popularising science (mostly in Polish):

  1. "Co z nami robi nieprzespany tydzień?", Małopolska Noc Naukowców, WFAIS UJ, Kraków, 30.09.2022, 29.09.2023 PREZENTACJA.
  2. "Nerwowa i krytyczna - sieć", Naukowe Czwartki, WFAIS UJ, Kraków, 2.12.2021 PREZENTACJA.
  3. naukawpolsce.pap.pl: Osoby wyspane odpoczywają… spokojniej niż niewyspane.
  4. naukawpolsce.pap.pl: Ślady niedoboru snu utrzymują się nawet siedem dni.
  5. everydayhealth.com: Sleep Debt Hampers Brain Function Up to a Week Later, Study Finds.
  6. healthline.com: It Can Take More Than a Week to Recover from Days of Sleep Deprivation.
  7. Poranek Radia TOK FM - Weekend, 2021-09-05.
  8. Other media coverage of the paper "Observing changes in human functioning ..."

Presentations, posters and files

  1. J. K. Ochab "Reinventing the Triangles: Rule of Thumb for Assessing Detectability," The 11th International Conference on Signal Image Technology & Internet Based Systems, Bangkok, 2015 POSTER.
  2. J.K. Ochab, "Dynamics of Maximal Entropy Random Walk and Generic Random Walk on Cayley trees," http://demonstrations.wolfram.com, January 31, 2012.
  3. J.K. Ochab, "Stationary states of Maximal Entropy Random Walk and Generic Random Walk on Cayley trees," http://demonstrations.wolfram.com, January 17, 2012.
  4. J. K. Ochab, Z. Burda, "Maximal-Entropy Random Walk, Centrality Measures and Communities," European Conference on Complex Systems, Brussels, 2012, POSTER.
  5. J. K. Ochab, Z. Burda, "Dynamics of Maximal Entropy Random Walk: solvable cases," European Conference on Complex Systems, Vienna, 2011, POSTER.
  6. J. Ochab, Z. Burda, W. Janke, H. Nagel, B. Waclaw, "Pair-Factorised Steady State model exhibits condensate''s growth on monolayers," 12th International NTZ-Workshop on New Developments in Computational Physics "CompPhys11", Leipzig, 2011PRESENTATION.
  7. J.K. Ochab, P.F. Góra, "Epidemic thresholds for a static and dynamic small-world network," Random geometry and applications, NORDITA, Stockholm, 2010PRESENTATION.
  8. Epidemic spread on a dynamic small-world network, ANIMATION (green pixels = susceptible, red = removed/recovered, yellow = infected/infectious; blue lines = long-range dynamic links that enhance the epidemic''s speed, range and toll).