Active and Self-Supervised Learning for Data Selection
Event Information
Event description
Smart data labeling - Igor Susmelj
About this Event
Outline
- Overview of recent methods in self-supervised learning with images and how to use them
- What can we do with the learned representations?
- Learn what is active-learning in the context of data selection
- Learn how we can combine the two approaches to improve the selection process
About the speaker:
Igor is a co-founder at Lightly – an ETH Spin-Off based in Zurich that is working on a novel data curation platform. Before, he worked for two years in the innovation lab of a financial institution as a software engineer. Igor holds a degree in electrical engineering from ETH Zurich. During his studies, he developed a lot of experience in machine learning and robotics and had multiple successful publications in the area of deep learning.
DataTalks.Club is the place to talk about data. Join our slack community!
Tags
Share with friends