In-memory Realization Of In-situ Few-shot Continual Learning With A Dynamically Evolving Explicit Memory

Abstract

Continually learning new classes from a few training examples without forgetting previous old classes demands a flexible architecture with an inevitably growing portion of storage, in which new examples and classes can be incrementally stored and efficiently retrieved. One viable architectural solution is to tightly couple a stationary deep neural network to a dynamically evolving explicit memory (EM). As the centerpiece of this architecture, we propose an EM unit that leverages energy-efficient in-memory compute (IMC) cores during the course of continual learning operations. We demonstrate for the first time how the EM unit can physically superpose multiple training examples, expand to accommodate unseen classes, and perform similarity search during inference, using operations on an IMC core based on phase-change memory (PCM). Specifically, the physical superposition of a few encoded training examples is realized via in-situ progressive crystallization of PCM devices. The classificati

Stats

• citations13
• S2 citations—
• github stars0
• HF likes0
• heat score8.60
• arxiv keykarunaratne2022in

Related papers

• Learning To Remember Rare Events (2017)0.00
• Towards Flexible Perception With Visual Memory (2024)0.00
• Learning With Memory-based Virtual Classes For Deep Metric Learning (2021)9.92
• CREAM: Continual Retrieval On Dynamic Streaming Corpora With Adaptive Soft Memory (2026)0.00
• Efficient And Reliable Vector Similarity Search Using Asymmetric Encoding With Nand-flash For Many-class Few-shot Learning (2024)2.26
• Semantically Constrained Memory Allocation (SCMA) For Embedding In Efficient Recommendation Systems (2021)0.00
• Retrieval-augmented Memory For Online Learning (2025)0.00
• SMEC: Rethinking Matryoshka Representation Learning For Retrieval Embedding Compression (2025)0.00