Recent searches
Search options
How robust are unsupervised representation learning methods (e.g. SSL) to distirbution shift compared to supervised learning?
𝐒𝐡𝐨𝐫𝐭 𝐚𝐧𝐬𝐰𝐞𝐫: Quite!
𝐋𝐨𝐧𝐠 𝐚𝐧𝐬𝐰𝐞𝐫: Our #ICLR2023 paper http://arxiv.org/pdf/2206.08871.pdf
Joint work with Imant Daunhawer & Amartya Sanyal @amartya
On realisitic domain generalisation datasets, supervised learning gains more momentum: on Camelyon17 it is second to SSL, but on FMoW it is the best performing method.
Still, considering that SSL requires no label during training, it's amazing how competitive it is! (3/n)