Feature-Based Echo-State Networks: A Step Towards Interpretability and Minimalism in Reservoir Computer
This work addresses interpretability and efficiency issues in reservoir computing for researchers and practitioners in time-series prediction, though it is incremental as it builds on the existing ESN paradigm.
The paper tackled the lack of interpretability and high computational complexity in traditional echo-state networks (ESNs) for time-series prediction by proposing a feature-based ESN (Feat-ESN) with smaller parallel reservoirs, resulting in outperformance over traditional ESNs with fewer reservoir nodes.
This paper proposes a novel and interpretable recurrent neural-network structure using the echo-state network (ESN) paradigm for time-series prediction. While the traditional ESNs perform well for dynamical systems prediction, it needs a large dynamic reservoir with increased computational complexity. It also lacks interpretability to discern contributions from different input combinations to the output. Here, a systematic reservoir architecture is developed using smaller parallel reservoirs driven by different input combinations, known as features, and then they are nonlinearly combined to produce the output. The resultant feature-based ESN (Feat-ESN) outperforms the traditional single-reservoir ESN with less reservoir nodes. The predictive capability of the proposed architecture is demonstrated on three systems: two synthetic datasets from chaotic dynamical systems and a set of real-time traffic data.