About me

I am a PhD Student at the Jagiellonian University. I work with GMUM research group, under the supervision of prof. Marek Śmieja. I am also a member of the IDEAS NCBR Computer Vision group, led by prof. Tomasz Trzciński.
I am interested in efficient deep learning, particularly in low-data regimes. I work on topics such as:

Self-supervised Learning - how to learn good model initializations from unlabeled data?
Meta-Learning - how to train models to learn novel tasks.
Few-Shot Learning - how to train models which can learn from very small amounts of data.

News

5.11.2024 - Augmentation-aware self-supervised learning with conditioned projector published in the Knowledge-based Systems journal!
3.10.2014 - Speaking at 33rd Polish Economic Forum at Karpacz about Self-supervised Learning, by invitation from the Ministry of Digital Affairs.
10.07.2014 - HyperMAML: Few-shot adaptation of deep models with hypernetworks published in the Neurocomputing journal!
1.06.2024 - A deep cut into Split Federated Self-supervised Learning accepted to ECML PKDD 2024. See you in Vilnius!
4.03.2024 - RegFlow: Probabilistic Flow-based Regression for Future Prediction accepted to ACIIDS 2024!
24.11.2023 - I've been awarded the NCN PRELUDIUM 22 grant for project "Improving the transferability of self-supervised learning models"!
read on
11.11.2023 - Augmentation-aware Self-Supervised Learning with Conditioned Projector accepted to Self-Supervised Learning - Theory and Practice workshop at NeurIPS 2023. See you in New Orleans!
24.10.2023 - Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual Learning accepted to WACV 2024!
13.10.2023 - Zero time waste in pre-trained early exit neural networks published in the Neural Networks journal!
2.10.2023 - I have started an internship at Computer Vision Center with prof. Joost van de Weijer. Until December, see you in Barcelona!