Learning Weight Uncertainty with Stochastic Gradient MCMC for Shape Classification
Authors:
- Chunyuan LiCLChunyuan Li
- Andrew StevensASAndrew Stevens
- Changyou ChenCCChangyou Chen
- Yunchen PuYPYunchen Pu
- Zhe GanZGZhe Gan
- Lawrence CarinLCLawrence Carin
- 1 June 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5666-5675
Abstract
Learning the representation of shape cues in 2D & 3D objects for recognition is a fundamental task in computer vision. Deep neural networks (DNNs) have shown promising performance on this task. Due to the large variability of shapes, accurate recognition relies on good estimates of model uncertainty, ignored in traditional training of DNNs, typically learned via stochastic optimization. This paper leverages recent advances in stochastic gradient Markov Chain Monte Carlo (SG-MCMC) to learn weight uncertainty in DNNs. It yields principled Bayesian interpretations for the commonly used Dropout/DropConnect techniques and incorporates them into the SG-MCMC framework. Extensive experiments on 2D & 3D shape datasets and various DNN models demonstrate the superiority of the proposed approach over stochastic optimization. Our approach yields higher recognition accuracy when used in conjunction with Dropout and Batch-Normalization.Keywords
This publication has 30 references indexed in Scilit:
- Bag of contour fragments for robust shape classificationPattern Recognition, 2014
- Intrinsic spatial pyramid matching for deformable 3D shape retrievalInternational Journal of Multimedia Information Retrieval, 2013
- A multiresolution descriptor for deformable 3D shape retrievalThe Visual Computer, 2013
- A survey of Bayesian predictive methods for model assessment, selection and comparisonStatistics Surveys, 2012
- Shape googleACM Transactions on Graphics, 2011
- A Concise and Provably Informative Multi‐Scale Signature Based on Heat DiffusionComputer Graphics Forum, 2009
- Spike and slab variable selection: Frequentist and Bayesian strategiesThe Annals of Statistics, 2005
- Human Upper Body Pose Estimation in Static Images2004
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- A Stochastic Approximation MethodThe Annals of Mathematical Statistics, 1951