14 November 2025
The Reservoir Computing (RC) paradigm is a unique approach for the design of untrained Recurrent Neural Networks (RNNs), popular for its computational efficiency. In this work, EMERGE partners from the University of Pisa introduce a novel class of untrained 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 the authors investigate diverse configurations for the temporal residual connections. The proposed approach is empirically assessed on time-series and pixel-level 1-D classification tasks. The group’s experimental results highlight the advantages of the proposed approach over other conventional RC models.
Read the paper in the link below.
More Information
