About the Tutorial

A topic that is constantly being discussed within Document Image Analysis (DIA) but also Machine Learning in general is the difficulties in reproducibility of scientific results. Many papers get published today where it is near impossible for the reader to independently reproduce and verify the reported numbers, due to various reasons.

In this tutorial we will introduce the topic of reproducibility and present general approaches to ensure reproducibility of research. In a practical session, we will discuss docker containers and how to set them up for one’s own code. Finally, we will offer a hands on session where we show how to create a research environment that is targeted towards reproducibility and can be easily shared with others.

In this tutorial we will provide and introduction and several hands-on sessions on existing solutions to reproduce DIA results.

About the program

The tutorial will be a half-day event. It comprises interactive pitch and demo sessions as well as hands-on experience.