Lift And Learn
Overview
The Lift and Learn method offers a transformative approach to model reduction in nonlinear dynamical systems, especially those not conforming to polynomial structures.
Lifting Process
This method introduces an auxiliary variable, effectively 'lifting' the system into a polynomial framework. This direct transformation contrasts with Koopman theory's approximation approach, streamlining the representation of complex dynamics.
Operator Inference
After lifting, the system is compatible with the Operator Inference scheme, facilitating the discovery of simplified models through data-driven techniques. This phase is instrumental in deriving interpretable and efficient reduced-order models from intricate dynamical systems.
For further details on lifting, please see lifting.