Constructive conditional normalizing flows
This work addresses conditional sampling in machine learning, offering incremental improvements in constructing normalizing flows for specific applications.
The paper tackles the problem of approximating diffeomorphisms and their pushforward measures for conditional sampling by constructing normalizing flows using neural networks with piecewise constant weights, providing explicit methods based on polar decomposition and probabilistic approaches for more regular maps.
Motivated by applications in conditional sampling, given a probability measure $μ$ and a diffeomorphism $φ$, we consider the problem of simultaneously approximating $φ$ and the pushforward $φ_{\#}μ$ by means of the flow of a continuity equation whose velocity field is a perceptron neural network with piecewise constant weights. We provide an explicit construction based on a polar-like decomposition of the Lagrange interpolant of $φ$. The latter involves a compressible component, given by the gradient of a particular convex function, which can be realized exactly, and an incompressible component, which -- after approximating via permutations -- can be implemented through shear flows intrinsic to the continuity equation. For more regular maps $φ$ -- such as the Knöthe-Rosenblatt rearrangement -- we provide an alternative, probabilistic construction inspired by the Maurey empirical method, in which the number of discontinuities in the weights doesn't scale inversely with the ambient dimension.