Skip to yearly menu bar Skip to main content

In-Person Poster presentation / top 25% paper

Learning with Stochastic Orders

Carles Domingo i Enrich · Yair Schiff · Youssef Mroueh

MH1-2-3-4 #160

Keywords: [ Choquet order ] [ stochastic order ] [ convex function ] [ statistical rates ] [ optimal transport ] [ input convex neural network ] [ image generation ] [ integral probability metric ] [ Theory ]


Learning high-dimensional distributions is often done with explicit likelihood modeling or implicit modeling via minimizing integral probability metrics (IPMs). In this paper, we expand this learning paradigm to stochastic orders, namely, the convex or Choquet order between probability measures. Towards this end, exploiting the relation between convex orders and optimal transport, we introduce the Choquet-Toland distance between probability measures, that can be used as a drop-in replacement for IPMs. We also introduce the Variational Dominance Criterion (VDC) to learn probability measures with dominance constraints, that encode the desired stochastic order between the learned measure and a known baseline. We analyze both quantities and show that they suffer from the curse of dimensionality and propose surrogates via input convex maxout networks (ICMNs), that enjoy parametric rates. We provide a min-max framework for learning with stochastic orders and validate it experimentally on synthetic and high-dimensional image generation, with promising results. Finally, our ICMNs class of convex functions and its derived Rademacher Complexity are of independent interest beyond their application in convex orders. Code to reproduce experimental results is available at

Chat is not available.