Self-Supervised learning in computer-vision

Hi Everyone, we're happy to announce our next hosted by Meta reality lab on self-supervised learning. The talk will be in Hebrew.

See you there!

Abstract:
Representation learning is arguably what deep-learning is all about. It has been shown that given enough manual annotation for a given task, effective visual representations can be learned and even be successfully transferred to related tasks with significantly fewer labels. Rapidly expanding to new tasks and data domains, manual labeling can no longer stand the pace. Accordingly, fascinating efforts and breakthroughs are being made for allowing effective representation learning without human supervision. This talk will cover the basics of self-supervised representation learning in computer-vision, as well as latest advancements. We will discuss the underlying assumptions, implications, and possible information-theoretic pitfalls of these ideas.
On the presenter:
Dr. Erez Farhan is a computer-vision applied researcher in the Colocation TLV team, Reality Labs, Meta.
His research in the team is focused on the problem of visual localization. Before coming to Meta, Erez worked in several Israeli start-ups as a researcher in the fields of Computer-Vision and Machine-Learning. Erez is a Phd graduate of the Ben-Gurion University.
Date: Tuesday 26/07/2022

Agenda:
1830 - 1900 gathering + food
1900-1910 About Reality labs and Meta
1910-2010 Presentation - Self-Supervised learning in computer-vision
Event address: Azrieli Sarona tower, Menachem Begin 121, Meta Offices
Instructions:

To participate ,please RSVP using this form: https://forms.gle/oX3t5zVs7E8mtWA98

After arriving to Meta's lobby at Azrieli Sarona tower you'll receive a tag and be escorted to the event room.