# Automated Variational Inference in Probabilistic Programming

@article{Wingate2013AutomatedVI, title={Automated Variational Inference in Probabilistic Programming}, author={David Wingate and Th{\'e}ophane Weber}, journal={ArXiv}, year={2013}, volume={abs/1301.1299} }

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… Expand

#### 100 Citations

Effective Monte Carlo Variational Inference for Binary-Variable Probabilistic Programs

- 2021

We propose a broadly applicable variational inference algorithm for probabilistic models with binary latent variables, using sampling to approximate expectations required for coordinate ascent… Expand

Symbolic Exact Inference for Discrete Probabilistic Programs

- Computer Science
- ArXiv
- 2019

This work provides a semantic and algorithmic foundation for efficient exact inference on discrete-valued finite-domain imperative probabilistic programs and shows that the inference approach is competitive with inference procedures specialized for Bayesian networks, thereby expanding the class of probabilism programs that can be practically analyzed. Expand

Towards verified stochastic variational inference for probabilistic programs

- Computer Science
- Proc. ACM Program. Lang.
- 2020

This paper analyses one of the most fundamental and versatile variational inference algorithms, called score estimator or REINFORCE, using tools from denotational semantics and program analysis, and formally expresses what this algorithm does on models denoted by programs, and exposes implicit assumptions made by the algorithm on the models. Expand

Automatic Variational Inference in Stan

- Computer Science, Mathematics
- NIPS
- 2015

An automatic variational inference algorithm, automatic differentiation Variational inference (ADVI), which is implemented in Stan, a probabilistic programming system and can be used on any model the authors write in Stan. Expand

Stochastically Differentiable Probabilistic Programs

- Computer Science, Mathematics
- ArXiv
- 2020

This work presents a novel approach to run inference efficiently and robustly in probabilistic programs with mixed support using stochastic gradient Markov Chain Monte Carlo family of algorithms and demonstrates that it outperforms existing composing inference baselines and works almost as well as inference in marginalized versions of the programs. Expand

Deep Amortized Inference for Probabilistic Programs

- Computer Science, Mathematics
- ArXiv
- 2016

A system for amortized inference in PPLs is proposed in the form of a parameterized guide program, which explores in detail the common machine learning pattern in which a 'local' model is specified by 'global' random values and used to generate independent observed data points; this gives rise to amortization local inference supporting global model learning. Expand

Automatic variational inference with cascading flows

- Computer Science, Mathematics
- ICML
- 2021

Cascading flows are introduced, a new family of variational programs that can be constructed automatically from an input probabilistic program and can also be amortized automatically that have much higher performance than both normalizing flows and ASVI in a large set of structured inference problems. Expand

Probabilistic Models with Deep Neural Networks

- Computer Science, Mathematics
- Entropy
- 2021

An overview of the main concepts, methods, and tools needed to use deep neural networks within a probabilistic modeling framework is provided. Expand

Reinforced Variational Inference

- Computer Science
- NIPS 2015
- 2015

The problem of learning approximate posterior distributions in variational inference (VI) onto the policy optimization problem in reinforcement learning (RL) is formally mapped, explaining this connection at two levels. Expand

Applications of Probabilistic Programming (Master's thesis, 2015)

- Computer Science
- ArXiv
- 2016

This thesis describes work on the learning of probabilistic program code given specifications, in particular program code of one-dimensional samplers; and the facilitation of sequential Monte Carlo inference with help of data-driven proposals. Expand

#### References

SHOWING 1-10 OF 47 REFERENCES

Stochastic variational inference

- Computer Science, Mathematics
- J. Mach. Learn. Res.
- 2013

Stochastic variational inference lets us apply complex Bayesian models to massive data sets, and it is shown that the Bayesian nonparametric topic model outperforms its parametric counterpart. Expand

A Stochastic approximation method for inference in probabilistic graphical models

- Computer Science, Mathematics
- NIPS
- 2009

A new algorithmic framework for inference in probabilistic models, and applies it to inference for latent Dirichlet allocation (LDA), which offers a principled means to exchange the variance of an importance sampling estimate for the bias incurred through variational approximation. Expand

Structured Variational Inference Procedures and their Realizations

- Computer Science
- AISTATS
- 2005

We describe and prove the convergence of several algorithms for approximate structured variational inference. We discuss the computation cost of these algorithms and describe their relationship to… Expand

Variational Program Inference

- Computer Science
- ArXiv
- 2010

The guide program is used as a proposal distribution in importance sampling to statistically prove lower bounds on the probability of the evidence and on the probabilities of a hypothesis and the evidence. Expand

Natural Conjugate Gradient in Variational Inference

- Mathematics, Computer Science
- ICONIP
- 2007

This work proposes using the geometry of the variational approximating distribution instead to speed up a conjugate gradient method for variational learning and inference, and shows significant speedups over alternative learning algorithms. Expand

An Introduction to Variational Methods for Graphical Models

- Mathematics, Computer Science
- Machine Learning
- 2004

This paper presents a tutorial introduction to the use of variational methods for inference and learning in graphical models (Bayesian networks and Markov random fields), and describes a general framework for generating variational transformations based on convex duality. Expand

Structured Variational Distributions in VIBES

- Computer Science, Mathematics
- AISTATS
- 2003

This paper presents an extension of VIBES in which the variational posterior distribution corresponds to a sub-graph of the full probabilistic model, which can produce much closer approximations to the true posterior distribution. Expand

Variational algorithms for approximate Bayesian inference

- Computer Science
- 2003

A unified variational Bayesian (VB) framework which approximates computations in models with latent variables using a lower bound on the marginal likelihood and is compared to other methods including sampling, Cheeseman-Stutz, and asymptotic approximations such as BIC. Expand

Hybrid Variational/Gibbs Collapsed Inference in Topic Models

- Computer Science, Mathematics
- UAI
- 2008

This paper proposes a hybrid algorithm that combines the best of both worlds: it samples very small counts and applies variational updates to large counts and is shown to significantly improve test-set perplexity relative to variational inference at no computational cost. Expand

Report on the probabilistic language scheme

- Computer Science
- DLS '07
- 2007

Probabilistic Scheme is presented, an embedding of probabilistic computation into Scheme that gives programmers an expressive language for implementing modular Probabilistic models that integrate naturally with the rest of Scheme. Expand