I'm pursuing PhD in AI at the Jagiellonian University, I live in Kraków, Poland.

Click on the relevant icon to find out more about myself.

Work

More info research page. You can find out more about my on-going projects by contacting me. Here I list my most recent projects.

My academic interests

I am into numerous areas of machine learning and mathematics, including (but not limitted to)

Bayesian Learning

Building models to comprehend the notion of uncertainty.

  • GFlowNets
  • Scalable Probabilistic
  • Variational Gradient Descent
  • Training with hypernetworks
  • Calibrated Meta-Learning

Continual Learning

Learning from non-stationary, continually changing data.

  • Links with Bayesian
  • Change of distribution
  • Buffor-free-based methods
  • Energy-based models in CL
  • Visualization in CL

Optimization

Designing universal approaches to reduce the computations.

  • Break-Even Point
  • Locality of Neurons.
  • Non-gradient-based training
  • Second-order information
  • Optimal Transport

Geometry and Topology

Understanding the structure and relation of underlying spaces.

  • Topology in ML
  • Symmetric structures
  • Loss landscape
  • Invariance in Neural Nets
  • Informative augmentations
Image

I am a very positive thinker, and I think that is what helps me the most in difficult moments.

Mateusz Pyla Roger Federer

Image

Do not lord it over the group which is in your charge, but be an example for the flock.

— 1 Peter 5, 3.

Image

It is all right letting yourself go, as long as you can get yourself back.

— The Rolling Stones