Balanced Population Stochastic Variational Inference

نویسندگان

  • Cheng Zhang
  • Stephan Mandt
  • Hedvig Kjellström
چکیده

Latent variable models are important tools to infer the underlying structure of a set of data. When we condition on observed data in Bayesian inference, we implicitly assume that the modeling assumptions are true and that the data can be considered a representative draw from the model. However, realistic data rarely agrees with these modeling assumptions. Especially when the observed data is highly imbalanced, inference will commonly result in redundant latent structures representing highly populated data clusters and miss the information contained in the scarce clusters. In this work, we propose a principled and scalable way to handle imbalanced data. We call our approach Balanced Population Stochastic Variational Inference (BP-SVI). Building on SVI, we use the Determinantal Point Process (DPP) to draw diversified mini-batches from the observed data set under the assumption that the observed data are an imbalanced realization of a true population. We show that this results in a more representative latent representation. From a theoretical side, our approach can be considered an instance of population Bayesian methods which were recently proposed for Bayesian inference on streams. While BP-SVI is applicable to a broad class of latent variable models, here we demonstrate how to use BP-SVI on Latent Dirichlet Allocation (LDA) as an example. Experiments are performed on both synthetic data and real-life news data showing clear improved performance on latent structure learning.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Adaptively Setting the Learning Rate in Stochastic Variational Inference

Stochastic variational inference is a promising method for fitting large-scale probabilistic models with hidden structures. Different from traditional stochastic learning, stochastic variational inference uses the natural gradient, which is particularly efficient for computing probabilistic distributions. One of the issues in stochastic variational inference is to set an appropriate learning ra...

متن کامل

Scalable Inference Algorithms for Clustering Large Networks

Clustering is an important task in network analysis, with applications in fields such as biology and the social sciences. We present a novel inference algorithm for the Stochastic Block Model (SBM), a well known network clustering model. Previous inference in this model typically utilizes Markov Chain Monte Carlo or Variational Bayes, but our method is the first to utilize Stochastic Variationa...

متن کامل

Variational Inference on Deep Exponential Family by using Variational Inferences on Conjugate Models

In this paper, we propose a new variational inference method for deep exponentialfamily (DEF) models. Our method converts non-conjugate factors in a DEF model to easy-to-compute conjugate exponential-family messages. This enables local and modular updates similar to variational message passing, as well as stochastic natural-gradient updates similar to stochastic variational inference. Such upda...

متن کامل

Truncation-free Stochastic Variational Inference for Bayesian Nonparametric Models

We present a truncation-free stochastic variational inference algorithm for Bayesian nonparametric models. While traditional variational inference algorithms require truncations for the model or the variational distribution, our method adapts model complexity on the fly. We studied our method with Dirichlet process mixture models and hierarchical Dirichlet process topic models on two large data...

متن کامل

Stochastic Variational Inference with Gradient Linearization

Variational inference has experienced a recent surge in popularity owing to stochastic approaches, which have yielded practical tools for a wide range of model classes. A key benefit is that stochastic variational inference obviates the tedious process of deriving analytical expressions for closed-form variable updates. Instead, one simply needs to derive the gradient of the log-posterior, whic...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2016