InferPy: Probabilistic modeling with deep neural networks made easy

Cózar, Javier; Cabañas, Rafael; Salmerón, Antonio; Masegosa, Andrés R.

doi:10.1016/j.neucom.2020.07.117

Cited by 2 publications

(4 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the main reasons for the wide adoption of deep learning has been the availability of (open-source) software tools containing robust and well-tested implementations of the main building blocks for defining and learning DNNs [ 34 , 35 ]. Recently, a new wave of software tools have appeared, building on top of these deep learning frameworks in order to accommodate modern probabilistic models containing DNNs [ 29 , 30 , 31 , 32 , 33 , 96 , 97 , 98 ]. These software tools usually fall under the umbrella term probabilistic programming languages (PPLs) [ 37 , 38 ], and support methods for learning and reasoning about complex probabilistic models.…”

Section: Probabilistic Programming Languagesmentioning

confidence: 99%

“…This framework is compatible with neural networks defined with Keras [99]. • InferPy [32,33] is a Python package built on top of Edward which focuses on the ease of use. It provides a compact and simple way to code probabilistic models with DNNs, at the expense of slightly reducing expressibility and flexibility.…”

Section: Probabilistic Programming Languagesmentioning

confidence: 99%

“…This new family of probabilistic models can encode objects like images, text, audio, and video probabilistically, thus bringing many of the recent advances produced by the deep learning community to the field of probabilistic modeling. The release of modern probabilistic programming languages [ 29 , 30 , 31 , 32 , 33 ] relying on well established deep learning engines like Tensorflow [ 34 ] and PyTorch [ 35 ] have also significantly contributed to the adoption of these powerful probabilistic modeling techniques.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Probabilistic Models with Deep Neural Networks

Masegosa

Cabañas

Langseth

et al. 2021

Entropy

Self Cite

View full text Add to dashboard Cite

Recent advances in statistical inference have significantly expanded the toolbox of probabilistic modeling. Historically, probabilistic modeling has been constrained to very restricted model classes, where exact or approximate probabilistic inference is feasible. However, developments in variational inference, a general form of approximate probabilistic inference that originated in statistical physics, have enabled probabilistic modeling to overcome these limitations: (i) Approximate probabilistic inference is now possible over a broad class of probabilistic models containing a large number of parameters, and (ii) scalable inference methods based on stochastic gradient descent and distributed computing engines allow probabilistic modeling to be applied to massive data sets. One important practical consequence of these advances is the possibility to include deep neural networks within probabilistic models, thereby capturing complex non-linear stochastic relationships between the random variables. These advances, in conjunction with the release of novel probabilistic modeling toolboxes, have greatly expanded the scope of applications of probabilistic models, and allowed the models to take advantage of the recent strides made by the deep learning community. In this paper, we provide an overview of the main concepts, methods, and tools needed to use deep neural networks within a probabilistic modeling framework.

show abstract

Section: Probabilistic Programming Languagesmentioning

confidence: 99%

Section: Probabilistic Programming Languagesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Probabilistic Models with Deep Neural Networks

Masegosa

Cabañas

Langseth

et al. 2021

Entropy

Self Cite

View full text Add to dashboard Cite

show abstract

“…ability of (open-source) software tools containing robust and well-tested implementations of the main building blocks for defining and learning DNNs (Abadi et al, 2015;Paszke et al, 2017). Recently, a new wave of software tools have appeared, building on top of these deep learning frameworks in order to accommodate modern probabilistic models containing DNNs (Tran et al, 2016;Cabañas et al, 2019;Cózar et al, 2019;Tran et al, 2018;Bingham et al, 2018). These software tools usually fall under the umbrella term probabilistic programming languages (PPLs) (Gordon et al, 2014;Ghahramani, 2015), and support methods for learning and reasoning about complex probabilistic models.…”

Section: Stochastic Computational Graphsmentioning

confidence: 99%

Probabilistic Models with Deep Neural Networks

Masegosa¹,

Cabañas²,

Langseth³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Recent advances in statistical inference have significantly expanded the toolbox of probabilistic modeling. Historically, probabilistic modeling has been constrained to very restricted model classes, where exact or approximate probabilistic inference is feasible. However, developments in variational inference, a general form of approximate probabilistic inference that originated in statistical physics, have enabled probabilistic modeling to overcome these limitations: (i) Approximate probabilistic inference is now possible over a broad class of probabilistic models containing a large number of parameters, and (ii) scalable inference methods based on stochastic gradient descent and distributed computing engines allow probabilistic modeling to be applied to massive data sets. One important practical consequence of these advances is the possibility to include deep neural networks within probabilistic models, thereby capturing complex non-linear stochastic relationships between the random variables. These advances, in conjunction with the release of novel probabilistic modeling toolboxes, have greatly expanded the scope of applications of probabilistic models, and allowed the models to take advantage of the recent strides made by the deep learning community. In this paper we review the main concepts, methods, and tools needed to use deep neural networks within a probabilistic modeling framework.

show abstract

InferPy: Probabilistic modeling with deep neural networks made easy

Cited by 2 publications

References 4 publications

Probabilistic Models with Deep Neural Networks

Probabilistic Models with Deep Neural Networks

Probabilistic Models with Deep Neural Networks

Contact Info

Product

Resources

About