Operator Surrogate Model
A class of data-driven methods for approximating operators between infinite-dimensional function spaces. By encoding, processing, and decoding input and output functions, these surrogate models can be trained once and then handle arbitrary new input functions at fixed cost during inference. They are mainly used for solving partial differential equations in scientific computing and engineering simulation.
Timeline (last 90 days)
Operator surrogate models are a class of finitely parameterized mappings used to approximate complex (usually nonlinear) operators between infinite-dimensional function spaces.
Traditional finite element methods require re-solving the system of equations for each new set of initial conditions, whereas a trained operator surrogate model can handle any new input function at inference time at fixed cost (one forward pass).
The theoretical core of operator surrogate models is to answer two questions: approximation error (whether the target operator can be approximated given enough parameters) and generalization error (whether it can generalize to unseen samples under finite training data).
All Facts (3)
Operator surrogate models are a class of finitely parameterized mappings used to approximate complex (usually nonlinear) operators between infinite-dimensional function spaces.
50%UnverifiedTraditional finite element methods require re-solving the system of equations for each new set of initial conditions, whereas a trained operator surrogate model can handle any new input function at inference time at fixed cost (one forward pass).
50%UnverifiedThe theoretical core of operator surrogate models is to answer two questions: approximation error (whether the target operator can be approximated given enough parameters) and generalization error (whether it can generalize to unseen samples under finite training data).
50%