Black Box Variational Inference

Abstract

Variational inference has become a widely used method to approximate posteriors in complex latent variables models. However, deriving a variational inference algorithm generally requires significant model-specific analysis, and these efforts can hinder and deter us from quickly developing and exploring a variety of models for a problem at hand. In this paper, we present a "black box" variational inference algorithm, one that can be quickly applied to many models with little additional derivation. Our method is based on a stochastic optimization of the variational objective where the noisy gradient is computed from Monte Carlo samples from the variational distribution. We develop a number of methods to reduce the variance of the gradient, always maintaining the criterion that we want to avoid difficult model-based derivations. We evaluate our method against the corresponding black box sampling based methods. We find that our method reaches better predictive likelihoods much faster than sampling methods. Finally, we demonstrate that Black Box Variational Inference lets us easily explore a wide space of models by quickly constructing and evaluating several models of longitudinal healthcare data.

References in corpus (1)

• Variational Bayesian Inference with Stochastic Search

Cited by in corpus (23)

• Linear dynamical neural population models through nonlinear embeddings
• Predictive Collective Variable Discovery with Deep Bayesian Models
• Stochastic gradient variational Bayes for gamma approximating distributions
• Kalman Gradient Descent: Adaptive Variance Reduction in Stochastic Optimization
• Embarrassingly Parallel Variational Inference in Nonconjugate Models
• Differentiable Particle Filtering without Modifying the Forward Pass
• Bayesian Adversarial Spheres: Bayesian Inference and Adversarial Examples in a Noiseless Setting
• Uncertainty aware audiovisual activity recognition using deep Bayesian variational inference
• On Reward Function for Survival
• Bayesian Neural Networks at Scale: A Performance Analysis and Pruning Study
• BAR: Bayesian Activity Recognition using variational inference
• Structured Factored Inference: A Framework for Automated Reasoning in Probabilistic Programming Languages
• VQ-DRAW: A Sequential Discrete VAE
• Probabilistic Circuits for Variational Inference in Discrete Graphical Models
• Generalized Hidden Parameter MDPs Transferable Model-based RL in a Handful of Trials
• Spreadsheet Probabilistic Programming
• Faster Stochastic Variational Inference using Proximal-Gradient Methods with General Divergence Functions
• Sampling Humans for Optimizing Preferences in Coloring Artwork
• Understanding Uncertainty in Bayesian Deep Learning
• Reinforcement Learning as Iterative and Amortised Inference
• Adaptive and Calibrated Ensemble Learning with Dependent Tail-free Process
• ADAVI: Automatic Dual Amortized Variational Inference Applied To Pyramidal Bayesian Models
• Variational Bayes Neural Network: Posterior Consistency, Classification Accuracy and Computational Challenges