Sebastian Cygert

I'm a postdoctoral researcher at IDEAS NCBR in Warsaw (Poland) and Assistant Professor at Gdańsk University of Technology (20%). At IDEAS NCBR I am a member of Continual Learning team within the Computer Vision group.

Before joining IDEAS I've worked in Amazon working on the Amazon Scout project (Tubingen, Germany) and Alexa Text-to-speech (Gdańsk, Poland). Before that I have worked at Machine Learning start-ups and have enjoyed the London FinTech scene. I have done my PhD at Gdańsk University of Technology and Masters at Warsaw University of Technology.

Email  /  Scholar  /  LinkedIn  /  Twitter  /  Github

profile photo


I'm interested in computer vision, deep learning and generative AI.
Most of the deep learning is about learning "good" representations. My research is about finding ways 1) to obtain robust representations, 2) to adapt them continually (e.g., after initial deployment) and 3) to share them across different models.

I am also interested in computational efficiency of ML models and physical simulations. Back in the old days I was doing low-level CUDA programing (e.g., for relative hydrodynamics simulations).

I am always looking for the motivated students, you can contact me by email.


3 Papers accepted to ECCV 2024!

We received NCN (Sonata) founding for our research on test-time adaptation, where I will be the main co-investigator (PI = B. Twardowski)!

Paper accepted to ICLR 2024


Improving platelet‐RNA‐based diagnostics: a comparative analysis of machine learning models for cancer detection and multiclass classification
Maksym Jopek, Krzysztof Pastuszak, Michał Sieczczyński, Sebastian Cygert, Anna Żaczek, Matthew T Rondina, Anna Supernat
Molecular Oncology, 2024

Divide and not forget: Ensemble of selectively trained experts in Continual Learning
Grzegorz Rypeść, Sebastian Cygert, Valeriya Khan, Tomasz Trzciński, Bartosz Zieliński, Bartłomiej Twardowski
ICLR, 2024

Looking through the past: better knowledge retention for generative replay in continual learning
Valeriya Khan, Sebastian Cygert, Kamil Deja, Tomasz Trzciński, Bartłomiej Twardowski
IEEE Access, 2024

Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual Learning
Filip Szatkowski, Mateusz Pyla, Marcin Przewięźlikowski, Sebastian Cygert, Bartłomiej Twardowski, Tomasz Trzciński
WACV, 2024

AR-TTA: A Simple Method for Real-World Continual Test-Time Adaptation
Damian Sójka, Sebastian Cygert, Bartłomiej Twardowski, Tomasz Trzciński
International Conference on Computer Vision (ICCV), CLVision workshop, 2023

Creating new voices using normalizing flows
Piotr Bilinski, Thomas Merritt, Abdelhamid Ezzerg, Kamil Pokora, Sebastian Cygert, Kayoko Yanagisawa, Roberto Barra-Chicote, Daniel Korzekwa
Interspeech, 2022

Robust object detection with multi-input multi-output faster R-CNN
Sebastian Cygert, Andrzej Czyżewski
International Conference on Image Analysis and Processing (ICIAP), 2022

Closer look at the uncertainty estimation in semantic segmentation under distributional shift
Sebastian Cygert, Bartłomiej Wróblewski, Karol Woźniak, Radosław Słowiński, Andrzej Czyżewski
IEEE International Joint Conference on Neural Networks (IJCNN), 2021

Robustness in Compressed Neural Networks for Object Detection
Sebastian Cygert, Andrzej Czyżewski
IEEE International Joint Conference on Neural Networks (IJCNN), 2021

Towards cancer patients classification using liquid biopsy
Sebastian Cygert, Franciszek Górski, Piotr Juszczyk, Sebastian Lewalski, Krzysztof Pastuszak, Andrzej Czyżewski, Anna Supernat
Predictive Intelligence in Medicine: 4th International Workshop, Held in Conjunction with MICCAI, 2021

Toward robust pedestrian detection with data augmentation
Sebastian Cygert, Andrzej Czyżewski
IEEE Access, 2020

Shape-Based Pose Estimation of Robotic Surgical Instruments
Daniel Węsierski, Sebastian Cygert
CARE Workshop, Held in Conjunction with MICCAI, 2017
Best Paper Award, 2nd place

Optimizing the computation of a parallel 3D finite difference algorithm for graphics processing units
Joanna Porter‐Sobieraj, Sebastian Cygert, Daniel Kikoła, Jan Sikorski, Marcin Słodkowski
Concurrency and Computation: Practice and Experience, 2015

My presentations

ML in PL 2023: YT

IDEAS NCBR podcast: YT

Great materials

R. Hamming's amazing talk - "You and Your Research": YT

P. Winston's amazing talk on How To Speak: YT

Guidelines for machine learning PhD students from M. Heinonen: link

Jason Wei slides on doing AI research: slides

Writing papers: great slides from B. Freeman: slides ; great video from S.P. Jones YT

This website is based on the Jon Barron template source code.