First-order ANIL provably learns representations despite overparametrization
2024
Overparameterization
A regime in which a model has more adjustable degrees of freedom than are needed to represent or interpolate the target structure.
A regime in which a model has more adjustable degrees of freedom than are needed to represent or interpolate the target structure.
Its consequences depend on the parameterization, learning algorithm, data, and criterion used to compare solutions.