Skip to yearly menu bar Skip to main content


Neural Fine-Tuning Search for Few-Shot Learning

Panagiotis Eustratiadis · Ɓukasz Dudziak · Da Li · Timothy Hospedales

Halle B #182
[ ]
Thu 9 May 1:45 a.m. PDT — 3:45 a.m. PDT
Oral presentation: Oral 5C
Thu 9 May 1 a.m. PDT — 1:45 a.m. PDT


In few-shot recognition, a classifier that has been trained on one set of classes is required to rapidly adapt and generalize to a disjoint, novel set of classes. To that end, recent studies have shown the efficacy of fine-tuning with carefully-crafted adaptation architectures. However this raises the question of: How can one design the optimal adaptation strategy? In this paper, we study this question through the lens of neural architecture search (NAS). Given a pre-trained neural network, our algorithm discovers the optimal arrangement of adapters, which layers to keep frozen, and which to fine-tune. We demonstrate the generality of our NAS method by applying it to both residual networks and vision transformers and report state-of-the-art performance on Meta-Dataset and Meta-Album.

Chat is not available.