Bio-inspired artificial neural networks

Foundation for Polish Science (FNP)
TEAM-NET programme

Our Project

The artificial neural network model was created based on analogies to biological counterparts, such as a simplified model of the neuron or a system of retinal neurons. Due to the increasing complexity of tasks and problems with the development of effective methods for learning deep neural networks, solutions based on algebraic structures dominate Today, advanced approaches in machine learning such as deep learning show a number of undesirable features, such as forgetfulness, susceptibility to adversarial examples, the requirement for a large training set, and slow learning. Most of these features do not occur in biological systems, thus it would be beneficial to take an inspiration from them to help training artificial systems. The aim of the project is to analyze high-level behaviors of biological neural systems and to build innovative artificial models by proposing new paradigms of learning and new architectures of computational models.

Job offers

Inquiry offers for anti-discrimination training, more details here.

The Jagiellonian University will run six research groups: Cognitive group (leader Tadeusz Marek), Physics group (leader Maciej A. Nowak), Machine-learning group, Neuro group, BioDataScience group, InfoTech group.

Job offers for

Cognitive group:

Physics group:

Machine-learning group:

Neuro group:

BioDataScience group:

InfoTech group:

are open.

If you are interested in applying, please fill out the following recruitment form (DOC) or (PDF).

Specific recruitment rules are available here. Declaration on intellectual is available here. The form for processing personal data can be found here.

Business contacts

Through our expertise, we support Ardigen SA in providing innovative solutions using our experience in medical image analysis, drug design, and cheminformatics. As a result, we published many research papers at top conferences (NeurIPS, IJCAI, MIDL, MEDINFO) and journals (Journal of Cheminformatics, PloS one). We plan to further strengthen the collaboration with Ardigen by submitting grant proposals to apply the results of our research in the business.

As part of our collaboration with Tooploox, we run several research projects focused on 3D point representations, continuous flow architectures of neural networks and visual reasoning. The results of these works are submitted and published at top-tier conferences (NeurIPS, ICML, IJCAI). Together with Tooploox, as our industrial partner, we plan to deepen our collaboration by submitting future grant proposals and turning research results into commercially viable solutions.

The principle of equal opportunities

We apply the principle of equal opportunities and implement standards of equality in project management. The project participants underwent 8-hour anti-discrimination training. The anti-discrimination guide for project participants is available here.

Seminars Schedule

17th December

Maria Lewandowska, "Głębokie konwolucyjne sieci neuronowe do analizy skanów MRI"

10th December

Dyskusja na temat danych pochodzących od grupy kongnitywistycznej

3rd December

Anna Ceglarek i Bartosz Bohaterewicz, "Zastosowanie teorii grafów w analizach fMRI"

5th November

Group Workshop

11:00 - 12:00	Machine-learning group
12:10 - 13:10	Physics group
13:10 - 14:00	Lunch
14:00 - 15:00	Neuro group
15:10 - 16:10	Cognitive group
16:10 - 16:40	Coffee Break
16:40 - 18:40	Group Discussion

Localization: Room 1146 (first floor). Faculty of Mathematics and Computer Science, Łojasiewicza 6, 30-348 Kraków.

24th September

Maciej Wołczyk, Jacek Tabor, Marek Śmieja, "Biologically-Inspired Spatial Neural Networks"

10th September

Romuald Janik, "Explaining the Human Visual Brain 2019 competition and workshop"
Additional information: Paper you can find here.

30th July

Magda Gawłowska, "EEG 101"

9th July

Igor Podolak, "BioNN machine learning"

8th December

11:15 - 12:45

Rafał Bogacz: Propagacja wsteczna w mózgu