Task-agnostic Continual Learning via Growing Long-Term Memory Networks

Germán Kruszewski, Ionut Teodor Sorodoc, Tomas Mikolov

TL;DR: We introduce a continual learning setup based on language modelling where no explicit task segmentation signal is given and propose a neural network model with growing long term memory to tackle it.

Abstract:

As our experience shows, humans can learn and deploy a myriad of different skills to tackle the situations they encounter daily. Neural networks, in contrast, have a fixed memory capacity that prevents them from learning more than a few sets of skills before starting to forget them.
In this work, we make a step to bridge neural networks with human-like learning capabilities. For this, we propose a model with a growing and open-bounded memory capacity that can be accessed based on the model’s current demands. To test this system, we introduce a continual learning task based on language modelling where the model is exposed to multiple languages and domains in sequence, without providing any explicit signal on the type of input it is currently dealing with. The proposed system exhibits improved adaptation skills in that it can recover faster than comparable baselines after a switch in the input language or domain.

Keywords: growing, long term memory, continual learning, catastrophic forgetting