Semi-Supervised iNaturalist Challenge

Overview

This challenge is focused on learning from partially labeled data, a form of semi-supervised learning. This dataset is designed to expose some of the challenges encountered in a realistic setting, such as the fine-grained similarity between classes, significant class imbalance, and domain mismatch between the labeled and unlabeled data. Different from last year, this year's challenge includes more species from different kingdoms and a mix of in-class and out-of-class unlabeled data.

Competition

Start Date - 10th March 2021

End Date - 31st May 2021

Kaggle URL - https://www.kaggle.com/c/semi-inat-2021

Organizers

Jong-Chyi Su (UMass Amherst)

Acknowledgements

We thank the FGVC team, Kaggle team, iNaturalist, Grant van Horn, and Oisin Mac Aodha for the help.