One-shot learning for the long term: consolidation with an artificial hippocampal algorithm

TL;DR

This paper introduces an artificial hippocampal algorithm that enables one-shot learning with long-term consolidation in machine learning models, preventing forgetting and mimicking mammalian memory processes.

Contribution

It presents a novel integration of an artificial hippocampal algorithm with incremental learning models to achieve long-term one-shot learning without catastrophic forgetting.

Findings

AHA enables one-shot learning in ML models.

The system consolidates knowledge long-term without forgetting.

First use of CLS hippocampus model for memory consolidation in ML.

Abstract

Standard few-shot experiments involve learning to efficiently match previously unseen samples by class. We claim that few-shot learning should be long term, assimilating knowledge for the future, without forgetting previous concepts. In the mammalian brain, the hippocampus is understood to play a significant role in this process, by learning rapidly and consolidating knowledge to the neocortex incrementally over a short period. In this research we tested whether an artificial hippocampal algorithm (AHA), could be used with a conventional Machine Learning (ML) model that learns incrementally analogous to the neocortex, to achieve one-shot learning both short and long term. The results demonstrated that with the addition of AHA, the system could learn in one-shot and consolidate the knowledge for the long term without catastrophic forgetting. This study is one of the first examples of using a CLS model of hippocampus to consolidate memories, and it constitutes a step toward few-shot continual learning.