Modular Deep Learning
Sebastian Ruder
FEBRUARY 23, 2023
Computation Function We consider a neural network $f_theta$ as a composition of functions $f_{theta_1} odot f_{theta_2} odot ldots odot f_{theta_l}$, each with their own set of parameters $theta_i$. d) Hypernetwork: A small separate neural network generates modular parameters conditioned on metadata.
Let's personalize your content