# Complementary Learning for Overcoming Catastrophic Forgetting Using Experience Replay

## TL;DR

This paper proposes a generative experience replay method based on complementary learning systems theory to mitigate catastrophic forgetting in deep neural networks during sequential multitask learning.

## Contribution

It introduces a novel framework that learns a shared generative distribution in an embedding space to connect current and past tasks, reducing forgetting.

## Key findings

- The method effectively prevents catastrophic forgetting in experiments.
- Theoretically demonstrates shared distribution across tasks.
- Empirically outperforms existing continual learning approaches.

## Abstract

Despite huge success, deep networks are unable to learn effectively in sequential multitask learning settings as they forget the past learned tasks after learning new tasks. Inspired from complementary learning systems theory, we address this challenge by learning a generative model that couples the current task to the past learned tasks through a discriminative embedding space. We learn an abstract level generative distribution in the embedding that allows the generation of data points to represent the experience. We sample from this distribution and utilize experience replay to avoid forgetting and simultaneously accumulate new knowledge to the abstract distribution in order to couple the current task with past experience. We demonstrate theoretically and empirically that our framework learns a distribution in the embedding that is shared across all task and as a result tackles catastrophic forgetting.