Automatic Differentiation of Programs with Discrete Randomness

Gaurav Arya · Moritz Schauer · Frank Schäfer · Christopher Rackauckas

Hall J #507

Keywords: [ reparameterization trick ] [ Stochastic Methods ] [ chain rule ] [ discrete randomness ] [ gradient based inference ] [ automatic differentiation ] [ differentiable stochastic programming ] [ compositionality ]

[ Abstract ]
[ Paper [ Poster [ OpenReview
Thu 1 Dec 9 a.m. PST — 11 a.m. PST

Abstract: Automatic differentiation (AD), a technique for constructing new programs which compute the derivative of an original program, has become ubiquitous throughout scientific computing and deep learning due to the improved performance afforded by gradient-based optimization. However, AD systems have been restricted to the subset of programs that have a continuous dependence on parameters. Programs that have discrete stochastic behaviors governed by distribution parameters, such as flipping a coin with probability $p$ of being heads, pose a challenge to these systems because the connection between the result (heads vs tails) and the parameters ($p$) is fundamentally discrete. In this paper we develop a new reparameterization-based methodology that allows for generating programs whose expectation is the derivative of the expectation of the original program. We showcase how this method gives an unbiased and low-variance estimator which is as automated as traditional AD mechanisms. We demonstrate unbiased forward-mode AD of discrete-time Markov chains, agent-based models such as Conway's Game of Life, and unbiased reverse-mode AD of a particle filter. Our code package is available at

Chat is not available.