Continual Learning with Echo State Networks
This work addresses the problem of continual learning for sequential patterns in fixed recurrent networks, which is incremental as it adapts existing CL strategies to ESNs.
The paper tackled catastrophic forgetting in Echo State Networks (ESNs) for continual learning, showing that ESNs are a promising model for this task and enabling their use in streaming scenarios.
Continual Learning (CL) refers to a learning setup where data is non stationary and the model has to learn without forgetting existing knowledge. The study of CL for sequential patterns revolves around trained recurrent networks. In this work, instead, we introduce CL in the context of Echo State Networks (ESNs), where the recurrent component is kept fixed. We provide the first evaluation of catastrophic forgetting in ESNs and we highlight the benefits in using CL strategies which are not applicable to trained recurrent models. Our results confirm the ESN as a promising model for CL and open to its use in streaming scenarios.