CAIML #19 - Why Relational Learning Matters

CAIML #19 will be an online meetup and it will happen on July 13, 2021, 18:30. We will have a talk by Sören Nikolaus from getML (https://getml.com/) on "Why relational learning matters" and a Q&A with Sören after the talk:

Agenda
18h30: Start of the meetup via MS Teams and short introduction
18h40: Sören's talk
19h10: Collect questions for Q&A with Sören via sli.do
19h15: 15 minutes Q&A

Sören Nikolaus is a Data Science Developer Advocate at getML. Sören mediates between customers, data scientists, and getML’s development team and advises applications of getML’s core technologies in various industries, including transportation, health, and finance. Sören is a trained economist (Osnabrück, Vienna, Leipzig) with hands-on experience in a variety of scientific fields, such as finance, neuroscience, and engineering.

-----

Why relational learning matters

Every machine learning algorithm expects its inputs to be structured as flat attribute-value tables. The problem: raw data is more complex than that. Convolutional neural nets introduced feature learning on images and text, the key ingredient to the deep learning revolution. Yet, in spite of the recent emergence of AutoML, for relational data, feature engineering is still mainly done by hand or using very simple brute force methods. Relational learning holds the potential to change that. In this talk, Sören Nikolaus, Data Science Developer Advocate at getML, will provide you with an overview of the current state of automated feature engineering and will introduce you to the latest advances in supervised feature learning on relational data.

-----