Łukasz Rączkowski

TL;DR: This work proposes an accurate, reliable and active (ARA) image classification framework and introduces a new Bayesian Convolutional Neural Network (ARA-CNN) for classifying histopathological images of colorectal cancer, which achieves exceptional classification accuracy, outperforming other models trained on the same dataset.

TL;DR: This work proposes an accurate, reliable and active (ARA) image classification framework and introduces a new Bayesian Convolutional Neural Network (ARA-CNN) for classifying histopathological images of colorectal cancer, which achieves exceptional classification accuracy, outperforming other models trained on the same dataset.

TL;DR: A small content-based visual recommendation project built as part of the Allegro online marketplace platform that extracted relevant data only from images, as they are inherently better at capturing visual attributes than textual offer descriptions.

TL;DR: This compendium is meant to serve as a guideline for established researchers, newcomers and students alike, highlighting interesting and rewarding problems in 'Single Cell Data Science' for the coming years.

TL;DR: In this paper , a probabilistic model, called Celloscope, was proposed for cell type deconvolution from spatial transcriptomics data, which utilizes established prior knowledge on marker genes for cell-type decoding.