Learning to Compare: Relation Network for Few-Shot Learning

• Published in: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 1199 - 1208
• Main Authors: Sung, Flood, Yang, Yongxin, Zhang, Li, Xiang, Tao, Torr, Philip H.S., Hospedales, Timothy M.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2018
• Subjects: Computer architecture; Image recognition; Measurement; Recurrent neural networks; Task analysis; Training

We present a conceptually simple, flexible, and general framework for few-shot learning, where a classifier must learn to recognise new classes given only few examples from each. Our method, called the Relation Network (RN), is trained end-to-end from scratch. During meta-learning, it learns to lear...