<ul><li>Reservoir Computing is a machine learning framework that can generalize to unexplored regions of state space without explicit structural priors.</li><li>A multiple-trajectory training scheme for reservoir computers supports training across disjoint time series, enabling effective use of available training data.</li><li>Reservoir Computing can achieve out-of-domain generalization by capturing system behavior in entirely unobserved basins of attraction.</li></ul>

Learning Beyond Experience: Generalizing to Unseen State Space with Reservoir Computing

Discover more