Automated Variational Inference in Probabilistic Programming

By David Wingate, Theophane Weber. 2013

152 citations recorded by the source corpus.

We present a new algorithm for approximate inference in probabilistic programs, based on a stochastic gradient for variational programs. This method is efficient without restrictions on the probabilistic program; it is particularly practical for distributions which are not analytically tractable, including highly structured distributions that arise in probabilistic programs. We show how to automatically derive mean-field probabilistic programs and optimize them, and demonstrate that our perspective improves inference efficiency over other algorithms.

Subfields: Artificial Intelligence, Ocean Engineering, Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Ocean Engineering

In Probabilistic & Bayesian Methods

Citations in the corpus

Variational Inference with Normalizing Flows (2015, background): “An important advantage of stochastic backpropagation is that, for models with continuous latent variables, it has the lowest variance among competing estimators.”
Gradient Estimation Using Stochastic Computation Graphs (2015, methodology): “The score function method is used to perform variational inference in general models (in the context of probabilistic programming) in Wingate and Weber [27], and similarly in Ranganath et al.”
An Introduction to Probabilistic Programming (2018, methodology): “Black-box variational inference (BBVI) (Wingate and Weber, 2013; Ranganath et al., 2014) optimizes L(λ) by performing gradient updates using a noisy estimate of the gradient ∇̂L(λ)”
Design and Implementation of Probabilistic Programming Language Anglican (2016, background): “Implementations of black-box variational inference (Wingate and Weber 2013; van de Meent et al. 2016) associate with random variables the learned parameters of variational posterior distribution.”
Black Box Variational Inference (2013, methodology): “This demonstrates the ease with which we can now consider models generally outside the realm of variational methods.”
Variational Inference: A Review for Statisticians (2016, background): “This first appeared independently in several articles (Ji, Shen, and West 2010; Nott et al. 2012; Paisley, Blei, and Jordan 2012; Wingate and Weber 2013).”
Automatic Differentiation Variational Inference (2016, background): “Wingate and Weber (2013) study a more general setting, where the variational approximation itself is a probabilistic program.”
The Generalized Reparameterization Gradient (2016, background): “…Paisley et al., 2012; Ranganath et al., 2014; Titsias and Lázaro-Gredilla, 2014), deep neural networks (Neal, 1992; Hin-ton et al., 1995; Mnih and Gregor, 2014; Kingma and Welling, 2014), and probabilistic programming (Wingate and Weber, 2013; Kucukelbir et al., 2015; van de Meent et al., 2016).”
REBAR: Low-variance, unbiased gradient estimates for discrete latent variable models (2017, methodology): “All of these involve learned baselines: from simple scalar baselines (Wingate & Weber, 2013; Ranganath et al., 2014) to nonlinear input-dependent baselines (Mnih & Gregor, 2014).”
Venture: a higher-order probabilistic programming platform with programmable inference (2014, methodology): “The idea of global mean-field inference in probabilistic programs via stochastic optimization, implemented using repeated re-simulation of the program, was first proposed in (Wingate and Weber, 2013).”
Deep Exponential Families (2014, methodology): “Thus we use more recent “black box” variational inference techniques that step around computing this expectation [40, 34, 27].”
Black-Box Policy Search with Probabilistic Programs (2015, methodology): “In this paper we develop an approach to policy learning based on black box variational inference (BBVI) [Ranganath et al., 2014, Wingate and Weber, 2013], a technique for variational approximation of the posterior in Bayesian models.”
Efficient Probabilistic Inference in the Quest for Physics Beyond the Standard Model (2018, methodology): “Recent developments in inference algorithms, such as variational methods [30], extensions that combine deep learning [31, 8], MCMC samplers based on physical dynamics such as the No-U-Turn Sampler (NUTS) [32] and Stochastic Gradient Langevin Dynamics [33], and methods that use deep neural networks t”
Deep Amortized Inference for Probabilistic Programs (2016, methodology): “Several existing general-purpose variational inference systems use this scheme [32, 18].”
Learning to Pass Expectation Propagation Messages (2013, unknown): “Also of note is [16], which applies mean field-like variational inference to general program specifications.”
Learning Stochastic Inverses (2013, background): “…p(h|d); vision is commonly understood as inverse graphics (Horn, 1977) and, more recently, as inverse physics (Sanborn et al., 2013; Watanabe and Shimojo, 2001); and conditional inference in probabilistic programs can be described as “running a program backwards” (e.g., Wingate and Weber, 2013).”
Reparameterization Gradient for Non-differentiable Models (2018, background): “Also, a variety of control variates for the standard score estimator [35, 27, 36, 28] for the gradients of variational objectives have been developed [28, 7, 8, 34, 6, 23], some of which use biased yet differentiable control variates such that the reparameterization trick can be used to correct the ”