One Proposal for Every Margin: Zero-Shot Amortized Sequential Importance Sampling for Binary Matrices
The proposed MarginFlow framework is a framework that turns the design of the proposal into a learning problem and amortizes it across margins by exploiting their self-similarity.