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

Learning a new concept from one example is a superior function of the humanbrain and it is drawing attention in the field of machine learning as aone-shot learning task. In this paper, we propose one of the simplest methodsfor this task with a nonparametric weight imprinting, named Direct ONE-shotlearning (DONE). DONE adds new classes to a pretrained deep neural network(DNN) classifier with neither training optimization nor pretrained-DNNmodification. DONE is inspired by Hebbian theory and directly uses the neuralactivity input of the final dense layer obtained from data that belongs to thenew additional class as the synaptic weight with a newly-provided-output neuronfor the new class, transforming all statistical properties of the neuralactivity into those of synaptic weight by quantile normalization. DONE requiresjust one inference for learning a new concept and its procedure is simple,deterministic, not requiring parameter tuning and hyperparameters. DONEovercomes a severe problem of existing weight imprinting methods thatDNN-dependently interfere with the classification of original-class images. Theperformance of DONE depends entirely on the pretrained DNN model used as abackbone model, and we confirmed that DONE with current well-trained backbonemodels perform at a decent accuracy.

It's DONE: Direct ONE-shot learning with quantile weight imprinting