#machine learning
Apr 2026
On Dominant Manifolds in Reservoir Computing Networks
This work studies the emergence of low-dimensional dominant manifolds in the training of Reservoir Computing networks for temporal forecasting tasks and shows that the training data generate an invariant subspace of the trained reservoir, whose dimension equals the number of dominant modes.
Noa Kaplan, Alberto Padoan, A. Bizyaeva
· arXiv.org · 2 citations