Abstract:A memristive few‐shot learning is demonstrated using analog cosine similarities. During the implementation, a hardware–software co‐optimization block is introduced in the encoding to reduce the memory size. Additionally, the lifelong memory adaptation method is proposed for the first time. Results achieve a great reduction of the memristor cells and obtain record‐level accuracies on the Omniglot dataset. External‐memory‐based few‐shot learning has emerged as a promising solution to deal with the shortage of labeled samples in traditional deep neural networks (DNNs). The external memories here are used to perform explicit similarity calculations. Despite several encouraging schemes were proposed to merge data storage and similarity calculation using non‐volatile memories, two major challenges, including the long encoding bits and the degraded accuracy of hardware implementations, remain to be tackled. Herein, a cosine‐similarity‐based few‐shot learning method with analog memristors is designed. A regulation block in the embedding net is used as a hardware–software co‐optimization unit to reduce the encoding lengths of the embedding vectors. Then, a lifelong memory adaptation method, termed the online prototype augmentation, is proposed for the first time to promote the accuracy of extremely low‐shot learning problems. Results show that the encoding bits are 4–32 times smaller than popular schemes. Using the Omniglot dataset, record accuracy (98.3% and 97.3% for the 5‐way 5‐shot and 5‐way 1‐shot problems, respectively) can be obtained for the few‐shot learning with memristive cosine similarities. In addition, the online prototype augmentation method leads to at least ≈ 2% improvement in accuracy for 1‐shot learning, which is outstanding compared to prior studies.

Memristive Cosine‐Similarity‐Based Few‐Shot Learning with Lifelong Memory Adaptation

Ferroelectric FET Based In-Memory Computing for Few-Shot Learning.

In-memory Realization of In-situ Few-shot Continual Learning with a Dynamically Evolving Explicit Memory

Efficient and Reliable Vector Similarity Search Using Asymmetric Encoding with NAND-Flash for Many-Class Few-Shot Learning

Few-Shot Learning in Spiking Neural Networks by Multi-Timescale Optimization

Few-Shot Object Recognition from Machine-Labeled Web Images.

Ferroelectric Ternary Content-Addressable Memory for One-Shot Learning

Learning with Memory for Few-Shot Semantic Segmentation

ROA: A Rapid Learning Scheme for In-Situ Memristor Networks.

MMT: Cross Domain Few-Shot Learning Via Meta-Memory Transfer.

Few-shot Continual Learning: a Brain-inspired Approach

Few-Shot Self Reminder to Overcome Catastrophic Forgetting

A Neocognitron Based on Multi-objective Optimization for Few-Shot Learning

Memory-Augmented Relation Network for Few-Shot Learning

Diagnosing and Remedying Shot Sensitivity with Cosine Few-Shot Learners

In Situ Training of Feed-Forward and Recurrent Convolutional Memristor Networks

Convolutional networks with short-term memory effects

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

Learning Meta-class Memory for Few-Shot Semantic Segmentation

A Transfer-Based Few-Shot Classification Approach Via Masked Manifold Mixup and Fuzzy Memory Contrastive Learning.

Learning to focus: cascaded feature matching network for few-shot image recognition