Augment and Reduce: Stochastic Inference for Large Categorical Distributions

Abstract

Categorical distributions are ubiquitous in machine learning, e.g., in classification, language models, and recommendation systems. However, when the number of possible outcomes is very large, using categorical distributions becomes computationally expensive, as the complexity scales linearly with the number of outcomes. To address this problem, we propose augment and reduce (A&R), a method to alleviate the computational complexity. A&R uses two ideas: latent variable augmentation and stochastic variational inference. It maximizes a lower bound on the marginal likelihood of the data. Unlike existing methods which are specific to softmax, A&R is more general and is amenable to other categorical models, such as multinomial probit. On several large-scale classification problems, we show that A&R provides a tighter bound on the marginal likelihood and has better predictive performance than existing approaches.

11 pages, 2 figures

References in corpus (5)

• Sticking the Landing: Simple, Lower-Variance Gradient Estimators for Variational Inference
• Conditional Probability Tree Estimation Analysis and Algorithms
• Deep Networks With Large Output Spaces
• Unbiased scalable softmax optimization
• Fast Amortized Inference and Learning in Log-linear Models with Randomly Perturbed Nearest Neighbor Search

Cited by in corpus (3)

• Adversarial Contrastive Estimation
• Variational f-divergence Minimization
• Scalable Gaussian Process Classification with Additive Noise for Various Likelihoods