Hi all,
cheers Darrel
That is a fundamental question, and one that is increasingly relevant in the age of "citizen science". Personally I think a lot of smaller pictures <"can help create the bigger picture">............ but many of them are associated with only a small amount of labeled data. Even though each individual task cannot be meaningfully trained in isolation, one seeks to meta-learn across the tasks from past experiences by exploiting some similarities. We study a fundamental question of interest: When can abundant tasks with small data compensate for a lack of tasks with big data?........
cheers Darrel
Last edited: