Details
Deep Learning, Rare Labels
===========================
Recent AI and Deep Learning breakthroughs have achieved some amazing results in a variety of tasks. However, one common barrier for using deep learning to solve hard problems is the amount of labeled data needed to train a model.
This meetup will focus on the challenge of training computer vision models using small, rare and sometimes sparse labels.
=====
Agenda:
18:00 – 18:30 Registration, Gathering, Mingling, Pizza & Beer.
18:30 – 19:15 AI Inside: Data Science and Computer Vision at Intel.
Dr. Amitai Armon, Chief Data Scientist, Intel Advanced Analytics.
19:15 – 19:30 Break.
19:30 – 20:15 Introducing “MAFAT Challenge”: Fine-Grained Classification of Objects from High-Resolution Aerial Imagery.
Eran Dahan, Israel Ministry of Defense.
Eran will present an ongoing computer vision contest, in which MAFAT is publicizing a high-resolution aerial imagery tagged dataset. Participants’ goal is to classify different objects found in aerial imagery data. The classification includes fine-grained classification of subclasses and unique features. Prizes (and everlasting glory) will be awarded to the winners.
See contest details at:
https://rebrand.ly/codald71dd