Continual Learning of Multi-modal Dynamics with External Memory

TL;DR

This paper introduces a novel continual learning approach for multi-modal dynamical systems that uses mode descriptors and a Dirichlet Process prior to effectively handle emerging behaviors without prior mode knowledge.

Contribution

The paper proposes a new continual learning method that maintains mode descriptors with a Dirichlet Process prior, overcoming limitations of existing approaches in multi-modal dynamical environments.

Findings

Outperforms traditional parameter transfer methods in continual learning tasks.

Effectively handles emerging modes without prior mode labels.

Utilizes mode descriptors and Dirichlet Process for efficient memory management.

Abstract

We study the problem of fitting a model to a dynamical environment when new modes of behavior emerge sequentially. The learning model is aware when a new mode appears, but it cannot access the true modes of individual training sequences. The state-of-the-art continual learning approaches cannot handle this setup, because parameter transfer suffers from catastrophic interference and episodic memory design requires the knowledge of the ground-truth modes of sequences. We devise a novel continual learning method that overcomes both limitations by maintaining a \textit{descriptor} of the mode of an encountered sequence in a neural episodic memory. We employ a Dirichlet Process prior on the attention weights of the memory to foster efficient storage of the mode descriptors. Our method performs continual learning by transferring knowledge across tasks by retrieving the descriptors of similar modes of past tasks to the mode of a current sequence and feeding this descriptor into its transition kernel as control input. We observe the continual learning performance of our method to compare favorably to the mainstream parameter transfer approach.