Learning feed-forward one-shot learners

Conference item

One-shot learning is usually tackled by using generative models or discriminative embeddings. Discriminative methods based on deep learning, which are very effective in other learning scenarios, are ill-suited for one-shot learning as they need large amounts of training data. In this paper, we propose a method to learn the parameters of a deep model in one shot. We construct the learner as a second deep network, called a learnet, which predicts the parameters of a pupil network from a single exemplar. In this manner we obtain an efficient feed-forward one-shot learner, trained end-to-end by minimizing a one-shot classification objective in a learning to learn formulation. In order to make the construction feasible, we propose a number of factorizations of the parameters of the pupil network. We demonstrate encouraging results by learning characters from single exemplars in Omniglot, and by tracking visual objects from a single initial exemplar in the Visual Object Tracking benchmark.

Authors: Bertinetto, L; Henriques, J; Valmadre, J; Torr, P; Vedaldi, A

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Massachusetts Institute of Technology Press
• Host title: 29th Conference on Neural Information Processing Systems (NIPS 2016), Monday December 05 - December 10, 2016 Barcelona, Spain
• Journal: Advances in Neural Information Processing Systems
• Publication date: 2016-12-05
• Acceptance date: 2016-05-20
• ISSN: 1049-5258