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

Kazufumi Hosoda, Keigo Nishida, Shigeto Seno, Tomohiro Mashita, Hideki KASHIOKA, Izumi Ohzawa

Keywords: One-shot learning, Few-shot learning, Weight imprinting, Hebbian theory, ImageNet models, New class addition

Abstract: Learning a new concept from one example is a superior function of the human brain and it is drawing attention in the field of machine learning as a one-shot learning task. In this paper, we propose one of the simplest methods for this task with a nonparametric weight imprinting, named Direct ONE-shot learning (DONE). DONE adds new classes to a pretrained deep neural network (DNN) classifier with neither training optimization nor pretrained-DNN modification. DONE is inspired by Hebbian theory and directly uses the neural activity input of the final dense layer obtained from data that belongs to the new additional class as the synaptic weight with a newly-provided-output neuron for the new class, by transforming all statistical properties of the neural activity into those of synaptic weight. DONE requires just one inference for learning a new concept and its procedure is simple, deterministic, not requiring parameter tuning and hyperparameters. DONE overcomes a problem of existing weight imprinting methods that interfere with the classification of original-class images. The performance of DONE depends entirely on the pretrained DNN model used as a backbone model, and we confirmed that DONE with current well-trained backbone models perform at a decent accuracy.

TL;DR: The simplest one-shot learning method with nonparametric weight imprinting is proposed. It is inspired by Hebbian theory and can add new classes to pretrained DNN at practical-level accuracy without optimization or the backbone-DNN modification.