Human-like Forgetting Curves in Deep Neural Networks

TL;DR

This paper demonstrates that deep neural networks can exhibit human-like forgetting curves by modeling memory decay and using scheduled reviews, which improves continual learning and aligns with cognitive science principles.

Contribution

It introduces a quantitative framework for measuring information retention in neural networks and shows how scheduled reviews can mitigate catastrophic forgetting.

Findings

Neural networks exhibit human-like forgetting curves.

Scheduled reviews enhance knowledge retention in neural networks.

Alignment with human memory models informs continual learning strategies.

Abstract

This study bridges cognitive science and neural network design by examining whether artificial models exhibit human-like forgetting curves. Drawing upon Ebbinghaus' seminal work on memory decay and principles of spaced repetition, we propose a quantitative framework to measure information retention in neural networks. Our approach computes the recall probability by evaluating the similarity between a network's current hidden state and previously stored prototype representations. This retention metric facilitates the scheduling of review sessions, thereby mitigating catastrophic forgetting during deployment and enhancing training efficiency by prompting targeted reviews. Our experiments with Multi-Layer Perceptrons reveal human-like forgetting curves, with knowledge becoming increasingly robust through scheduled reviews. This alignment between neural network forgetting curves and established human memory models identifies neural networks as an architecture that naturally emulates human memory decay and can inform state-of-the-art continual learning algorithms.