Residual Reservoir Memory Networks
This work addresses a specific bottleneck in reservoir computing for researchers in time-series analysis, though it appears incremental as it builds on existing RC paradigms.
The authors tackled the problem of enhancing long-term memory in reservoir computing networks by introducing Residual Reservoir Memory Networks (ResRMNs), which combine linear and non-linear reservoirs with residual orthogonal connections, achieving better performance than conventional RC models on time-series and 1-D classification tasks.
We introduce a novel class of untrained Recurrent Neural Networks (RNNs) within the Reservoir Computing (RC) paradigm, called Residual Reservoir Memory Networks (ResRMNs). ResRMN combines a linear memory reservoir with a non-linear reservoir, where the latter is based on residual orthogonal connections along the temporal dimension for enhanced long-term propagation of the input. The resulting reservoir state dynamics are studied through the lens of linear stability analysis, and we investigate diverse configurations for the temporal residual connections. The proposed approach is empirically assessed on time-series and pixel-level 1-D classification tasks. Our experimental results highlight the advantages of the proposed approach over other conventional RC models.