23 February 2022 at 5 pm
HubHub Twin City, Bratislava & online

Do you care for Datasets as much as for code in your ML Project?

This time, we want to give you insight into how AI and machine learning methods are put to practical use at Innovatrics for training new biometric algorithms and in Swiss Re, one of the largest re-insurance companies in the world, to develop new, innovative products.

Machine learning (ML) helped us to solve such complex tasks as face recognition, however when it is done on a large scale it relies on many high-effort data management tasks. We will describe our winding road from images in folders and labels stored in CSV files to structured versioned datasets with reproducible data processing pipelines. Moreover we mention how we deal with data versioning / validation / quality evaluation, reproducible ML trainings and experiment management.




Marian Beszedes


Marian Beszedes, Head of Innovatrics’ R&D Center will describe the journey from images in folders and CSV files to structured versioned dataset with reproducible pipelines and statistical comparable profiles. He will also explain how we selected 3rd party tooling (dvc, gitlab, pandas, TF validation, HDF5) as well as why we have written some of the tools by ourselves. Moreover we mention our deep NN training workflow and experiment management platform (ClearML).Gutentor Simple Text


Lukas Csoka

Swiss Re

Lukas Csoka, the Head of Big Data Foundations at Swiss Re, will provide insights into how to tame really large, diverse data into producing meaningful results. Swiss Re uses these results to design new products that are innovative and dramatically different from the traditional ones — all thanks to proper use and analysis of spatial, photographic, satellite, language and other data.


Language: English

5:00 pm
5:10 pm
Presentation – Do you care for Datasets as much as for code in your ML Project ? – Marian Beszedes (Innovatrics)
5:40 pm
Discussion – How Swiss Re uses AI for analyzing big data – Lukas Csoka (Swiss Re)
6:10 pm
Live Q&A – _Slido #better_ai

Past Meetups