Search Results for author: Scott A. Sisson

Found 11 papers, 2 papers with code

Online Binary Space Partitioning Forests

no code implementations29 Feb 2020 Xuhui Fan, Bin Li, Scott A. Sisson

The Binary Space Partitioning-Tree~(BSP-Tree) process was recently proposed as an efficient strategy for space partitioning tasks.

Classification General Classification

Bayesian Nonparametric Space Partitions: A Survey

no code implementations26 Feb 2020 Xuhui Fan, Bin Li, Ling Luo, Scott A. Sisson

Bayesian nonparametric space partition (BNSP) models provide a variety of strategies for partitioning a $D$-dimensional space into a set of blocks.

Smoothing Graphons for Modelling Exchangeable Relational Data

no code implementations25 Feb 2020 Xuhui Fan, Yaqiong Li, Ling Chen, Bin Li, Scott A. Sisson

We initially propose the Integrated Smoothing Graphon (ISG) which introduces one smoothing parameter to the SBM graphon to generate continuous relational intensity values.

Link Prediction Stochastic Block Model

Recurrent Dirichlet Belief Networks for Interpretable Dynamic Relational Data Modelling

no code implementations24 Feb 2020 Yaqiong Li, Xuhui Fan, Ling Chen, Bin Li, Zheng Yu, Scott A. Sisson

In this work, we leverage its interpretable modelling architecture and propose a deep dynamic probabilistic framework -- the Recurrent Dirichlet Belief Network~(Recurrent-DBN) -- to study interpretable hidden structures from dynamic relational data.

Link Prediction

Logistic regression models for aggregated data

no code implementations9 Dec 2019 Tom Whitaker, Boris Beranger, Scott A. Sisson

Logistic regression models are a popular and effective method to predict the probability of categorical response data.

Classification Crop Classification +1

Efficient Bayesian synthetic likelihood with whitening transformations

no code implementations11 Sep 2019 Jacob W. Priddle, Scott A. Sisson, David T. Frazier, Christopher Drovandi

Bayesian synthetic likelihood (BSL) is a popular such method that approximates the likelihood function of the summary statistic with a known, tractable distribution -- typically Gaussian -- and then performs statistical inference using standard likelihood-based techniques.

Bayesian Inference

Vector operations for accelerating expensive Bayesian computations -- a tutorial guide

1 code implementation25 Feb 2019 David J. Warne, Scott A. Sisson, Christopher Drovandi

We illustrate the potential of SIMD for accelerating Bayesian computations and provide the reader with techniques for exploiting modern massively parallel processing environments using standard tools.

Distributed Computing

Variance reduction properties of the reparameterization trick

no code implementations27 Sep 2018 Ming Xu, Matias Quiroz, Robert Kohn, Scott A. Sisson

From this, we show that the marginal variances of the reparameterization gradient estimator are smaller than those of the score function gradient estimator.

Variational Inference

New models for symbolic data analysis

no code implementations11 Sep 2018 Boris Beranger, Huan Lin, Scott A. Sisson

We assume that, as with a standard statistical analysis, inference is required at the level of individual-level data.

Efficient data augmentation for multivariate probit models with panel data: An application to general practitioner decision-making about contraceptives

1 code implementation19 Jun 2018 Vincent Chin, David Gunawan, Denzil G. Fiebig, Robert Kohn, Scott A. Sisson

This article considers the problem of estimating a multivariate probit model in a panel data setting with emphasis on sampling a high-dimensional correlation matrix and improving the overall efficiency of the data augmentation approach.

Computation Applications Methodology

Blocking Collapsed Gibbs Sampler for Latent Dirichlet Allocation Models

no code implementations2 Aug 2016 Xin Zhang, Scott A. Sisson

In this article, we introduce a blocking scheme to the collapsed Gibbs sampler for the LDA model which can, with a theoretical guarantee, improve chain mixing efficiency.

Cannot find the paper you are looking for? You can Submit a new open access paper.