Title
Learning representations from high-level structures
Abstract
This presentation will discuss new breakthroughs in self-supervised representation learning that depart from traditional instance-based contrastive learning. Current methods in self-supervised learning heavily rely on contrasting samples, but this approach presents challenges when augmentations are not feasible within the domain, alter the problem, or when effective data augmentation strategies are unclear. To address these issues, this talk introduces the concept of high-level supervisory signals and advocates for supervised learning of higher-order structures. The speaker will also touch briefly on their research agenda and current topics to link them with Integreat, but the main focus will be on the novel results in self-supervised representation learning.
Bio
Professor in the Digital Signal Processing and Image Analysis (DSB) group, Section for Machine Learning, Department of Informatics, University of Oslo. Ramírez Rivera's main research area is representation learning. He is currently exploring the use of machine learning to describe and understand the world primarily through vision sources, but not limited to. Ramírez Rivera is interested in applying these representations to various Computer Vision tasks and utilizing different Machine Learning methods to extract them.
Practical
The speaker is presenting in person at the Blindern campus, Niels Henrik Abels building, 8th floor.
The Tuesdays seminars series are devoted to various topics relevant for the Integreat´s research focus. Presenters from the Integreat community and beyond have 40 minutes to present, followed by a group discussion.
Seminars are open for attendance for everybody.
For those unable to attend in person: https://uio.zoom.us/j/62535744013
Point of contacts:
- Organiser and moderator: Professor Arnoldo Frigessi, University of Oslo
- Integreat Administrative Leader Maria Dikova, University of Oslo