A Growing Long-term Episodic & Semantic Memory

TL;DR

This paper introduces a lifelong learning system that uses a fast, content-addressable memory to store an unbounded amount of episodic and semantic knowledge, overcoming traditional weight-bound limitations.

Contribution

The paper presents a novel lifelong learning architecture combining fixed-size neural weights with a non-differentiable memory for scalable knowledge storage across multiple domains.

Findings

Enables unbounded knowledge accumulation over time.

Supports transfer learning across domains.

Demonstrates effective episodic and semantic memory integration.

Abstract

The long-term memory of most connectionist systems lies entirely in the weights of the system. Since the number of weights is typically fixed, this bounds the total amount of knowledge that can be learned and stored. Though this is not normally a problem for a neural network designed for a specific task, such a bound is undesirable for a system that continually learns over an open range of domains. To address this, we describe a lifelong learning system that leverages a fast, though non-differentiable, content-addressable memory which can be exploited to encode both a long history of sequential episodic knowledge and semantic knowledge over many episodes for an unbounded number of domains. This opens the door for investigation into transfer learning, and leveraging prior knowledge that has been learned over a lifetime of experiences to new domains.