Deep Generative Dual Memory Network for Continual Learning

TLDR

Neural networks trained sequentially suffer catastrophic forgetting, limiting their ability to learn multiple tasks without joint training. The study aims to create a human‑memory inspired architecture that learns continuously from sequential tasks while preventing catastrophic forgetting. The proposed dual‑memory network emulates hippocampus and neocortex, uses generative replay for memory consolidation, and is evaluated experimentally to show improved retention on challenging tasks. The architecture exhibits mammalian‑like memory traits and offers insights into the link between sleep and learning.

Abstract

Despite advances in deep learning, neural networks can only learn multiple tasks when trained on them jointly. When tasks arrive sequentially, they lose performance on previously learnt tasks. This phenomenon called catastrophic forgetting is a fundamental challenge to overcome before neural networks can learn continually from incoming data. In this work, we derive inspiration from human memory to develop an architecture capable of learning continuously from sequentially incoming tasks, while averting catastrophic forgetting. Specifically, our contributions are: (i) a dual memory architecture emulating the complementary learning systems (hippocampus and the neocortex) in the human brain, (ii) memory consolidation via generative replay of past experiences, (iii) demonstrating advantages of generative replay and dual memories via experiments, and (iv) improved performance retention on challenging tasks even for low capacity models. Our architecture displays many characteristics of the mammalian memory and provides insights on the connection between sleep and learning.