English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Modern deep learning systems are increasingly deployed in situations such aspersonalization and federated learning where it is necessary to support i)learning on small amounts of data, and ii) communication efficient distributedtraining protocols. In this work, we develop FiLM Transfer (FiT) which fulfillsthese requirements in the image classification setting by combining ideas fromtransfer learning (fixed pretrained backbones and fine-tuned FiLM adapterlayers) and meta-learning (automatically configured Naive Bayes classifiers andepisodic training) to yield parameter efficient models with superiorclassification accuracy at low-shot. The resulting parameter efficiency is keyfor enabling few-shot learning, inexpensive model updates for personalization,and communication efficient federated learning. We experiment with FiT on awide range of downstream datasets and show that it achieves betterclassification accuracy than the leading Big Transfer (BiT) algorithm atlow-shot and achieves state-of-the art accuracy on the challenging VTAB-1kbenchmark, with fewer than 1% of the updateable parameters. Finally, wedemonstrate the parameter efficiency and superior accuracy of FiT indistributed low-shot applications including model personalization and federatedlearning where model update size is an important performance metric.

FiT: Parameter Efficient Few-shot Transfer Learning for Personalized and Federated Image Classification