Schedule & Accepted Papers

Confirmed Speakers


8.00 - 8.05 Opening remarks
8.05 - 8.25 Invited talk Alexander G. de G. Matthews (DeepMind) Gaussian Process Behaviour in Wide Deep Neural Networks
8.25 - 8.40 Contributed talk Tim G. J. Rudner, Florian Wenzel and Yarin Gal The Natural Neural Tangent Kernel: Neural Network Training Dynamics under Natural Gradient Descent
8.40 - 9.00 Invited talk Yingzhen Li (Microsoft Research) On estimating epistemic uncertainty (tentative)
9.00 - 9.15 Contributed talk Stanislav Fort, Huiyi Hu and Balaji Lakshminarayanan Deep Ensembles: A Loss Landscape Perspective
9.15 - 9.30 Poster spotlights
9.30 - 10.30 Discussion over coffee and poster session
10.30 - 10.50 Invited talk Andrew Gordon Wilson (NYU) Using Loss Surface Geometry for Scalable Bayesian Deep Learning
10.50 - 11.05 Contributed talk Abhishek Kumar and Ben Poole On Implicit Regularization in β-VAE
11.05 - 11.25 Invited talk Jasper Snoek (Google) Uncertainty under distributional shift
11.25 - 11.40 Contributed talk Sicong Huang, Alireza Makhzani, Yanshuai Cao and Roger Grosse Evaluating Lossy Compression Rates of Deep Generative Models
11.40 - 13.20 Lunch
13.20 - 13.40 Invited talk Chelsea Finn (Google Brain / Berkeley / Stanford) The Big Problem with Meta-Learning and How Bayesians Can Fix It
13.40 - 13.55 Contributed talk Riccardo Moriconi, Marc Peter Deisenroth and Senanayak Sesh Kumar Karri High-dimensional Bayesian optimization using low-dimensional feature spaces
13.55 - 14.15 Invited talk Roger Grosse (Toronto) Functional variational Bayesian neural networks
14.15 - 14.30 Contributed talk Sebastian Farquhar and Yarin Gal You Need Depth, Not Weight Correlations: Mean-field is Not a Restrictive Assumption in Variational Inference for Deep Networks
14.30 - 15.30 Discussion over coffee and poster session
15.30 - 15.45 Contributed talk Andrew Ross, Jianzhun Du, Yonadav Shavit and Finale Doshi-Velez Controlled Direct Effect Priors for Bayesian Neural Networks
15.45 - 16.05 Invited talk Debora Marks (Harvard Medical School / Broad Institute) Deep generative models for genetic variation and drug design
16.05 - 16.20 Contributed talk Mihaela Rosca, Michael Figurnov, Shakir Mohamed and Andriy Mnih Measure Valued Derivatives for Approximate Bayesian Inference
16.20 - 17.30 Panel Session Panellists:
Debora Marks
Jasper Snoek
Chelsea Finn
Andrew Gordon Wilson
Yingzhen Li
Roger Grosse
Alexander G. de G. Matthews
Moderator: Finale Doshi-Velez
17.30 - 19.30 Poster session

Accepted Abstracts

Authors Title
Andrew Y. K. Foong, David R. Burt, Yingzhen Li and Richard E. TurnerPathologies of Factorised Gaussian and MC Dropout Posteriors in Bayesian Neural Networks
Paulo Rauber, Aditya Ramesh andJürgen SchmidhuberRecurrent neural-linear posterior sampling for non-stationary bandits
Andreas Look and Melih KandemirDifferential Bayesian Neural Nets
Matt Benatan and Edward Pyzer-KnappFully Bayesian Recurrent Neural Networks for Safe Reinforcement Learning
Ruiyi Zhang, Changyou Chen, Zhe Gan, Zheng Wen and Lawrence CarinNested-Wasserstein Self-Imitation Learning for Sequence Generation
Marton Havasi, Jasper Snoek, Dustin Tran, Jonathan Gordon and Jose Miguel Hernandez-LobatoRefining the variational posterior through iterative optimization
Hlynur Jonsson,Giovanni Cherubiniand Evangelos EleftheriouConvergence of DNNs with mutual-information-based regularization
Fredrik K. Gustafsson, Martin Danelljan and Thomas B. SchönEvaluating Scalable Bayesian Deep Learning Methods for Robust Computer Vision
Angelos Filos, Sebastian Farquhar, Aidan N. Gomez, Tim G. J. Rudner, Zachary Kenton, Lewis Smith, Milad Alizadeh, Arnoud de Kroon and Yarin GalBenchmarking Bayesian Deep Learning withDiabetic Retinopathy Diagnosis
James Brofos, Rui Shu and Roy LedermanA Bias-Variance Decomposition for Bayesian Deep Learning
Nabeel Seedat and Christopher KananTowards calibrated and scalable uncertainty representations for neural networks
Otmane Sakhi, Stephen Bonner, David Rohde and Flavian VasileReconsidering analytical variational bounds for output layers of deep networks
Micha Livne,Kevin SwerskyandDavid FleetHigh Mutual Information in Representation Learning with Symmetric Variational Inference
Hao Fu, Chunyuan Li, Ke Bai, Jianfeng Gao and Lawrence CarinFlexible Text Modeling withSemi-Implicit Latent Representations
Suwen Lin, Martin Wistuba, Ambrish Rawat and Nitesh ChawlaNeural Tree Kernel Learning
Ivan Ustyuzhaninov, Ieva Kazlauskaite, Markus Kaiser, Erik Bodin, Carl Henrik Ek and Neill CampbellCompositional uncertainty in deep Gaussian processes
Adam D Cobb, Atilim Günes Baydin, Ivan Kiskin, Andrew Markham and Stephen RobertsSemi-separable Hamiltonian Monte Carlo for inference in Bayesian neural networks
Eric Nalisnick, Akihiro Matsukawa, Yee Whye Teh and Balaji LakshminarayananDetecting Out-of-Distribution Inputs to Deep Generative Models Using Typicality
Felix McGregor, Arnu Pretorius, Johan du Preez and Steve KroonStabilising priors for robust Bayesian deep learning
John Moberg, Lennart Svensson, Juliano Pinto and Henk WymeerschBayesian Linear Regression on Deep Representations
Bang An, Xuannan Dong and Changyou ChenRepulsive Bayesian Sampling for Diversified Attention Modeling
Colin White,Willie NeiswangerandYash SavaniDeep Uncertainty Estimation for Model-based Neural Architecture Search
Vincent Fortuin and Gunnar RätschDeep Mean Functions for Meta-Learning in Gaussian Processes
Anirudh Suresh and Srivatsan SrinivasanImproved Attentive Neural Processes
Patrick McClure, Nao Rho, John Lee, Jakub Kaczmarzyk, Charles Zheng, Satrajit Ghosh, Dylan Nielson, Adam Thomas, Peter Bandettini and Francisco PereiraImproving 3D Brain Segmentation using a Spike-and-Slab Bayesian Deep Neural Network
Tim R. Davidson, Jakub M. Tomczak and Efstratios GavvesIncreasing Expressivity of a Hyperspherical VAE
Gaurush Hiranandani, Sumeet Katariya,Nikhil RaoandKarthik SubbianOnline Bayesian Learning for E-Commerce Query Reformulation
Sanjeev Arora, Simon Du, Zhiyuan Li, Ruslan Salakhutdinov, Ruosong Wang and Dingli YuOn the Power of NTK on Small Data
Prithvijit ChakrabartyandSubhransu MajiThe Spectral Bias of the Deep Image Prior
Masha Itkina, Boris Ivanovic, Ransalu Senanayake, Mykel Kochenderfer and Marco PavoneEvidential Disambiguation of Latent Multimodality in Conditional Variational Autoencoders
Simone Rossi, Sébastien Marmin and Maurizio FilipponeEfficient Approximate Inference with Walsh-Hadamard Variational Inference
Didrik Nielsenand Ole WintherPixelCNN as a Single-Layer Flow
Changyong Oh, Kamil Adamczewski and Mijung ParkThe Radial and Directional Posteriors for Bayesian Deep Learning
William Harvey, Michael Teng andFrank WoodNear-Optimal Glimpse Sequences for Training Hard Attention Neural Networks
Matias Valdenegro-ToroDeep Sub-Ensembles for Fast Uncertainty Estimation in Image Classification
Stefano Peluchetti and Stefano FavaroNeural SDE - Information propagation through the lens of diffusion processes
Riccardo Moriconi, Marc Peter Deisenroth and Senanayak Sesh Kumar KarriHigh-dimensional Bayesian optimization using low-dimensional feature spaces
Ali Hebbal, Loic Brevault, Mathieu Balesdent, El-Ghazali Talbi and Nouredine MelabMulti-fidelity modeling using DGPs: Improvements and a generalization to varying input space dimensions
Wen Yao, Jun Zhang, Qiang Chang, Xiaozhou Zhu and Weien ZhouError Estimation of Sampling-free Uncertainty Propagation in Bayesian Neural Network with Simplified Covariance Matrix
Sebastian Farquhar and Yarin Gal"You Need Depth, Not Weight Correlations: Mean-field is Not a Restrictive Assumption in Variational Inference for Deep Networks"
Tianyu Cui,Pekka MarttinenandSamuel KaskiLearning Global Pairwise Interactions with Bayesian Neural Networks
Da Tang,Dawen Liang,Nicholas RuozziandTony JebaraLearning Correlated Latent Representations with Adaptive Priors
Vaclav Smidl, Jan Bim and Tomas PevnyOrthogonal Approximation of Marginal Likelihood of Generative Models
Augustin Prado, Ravinath Kausik and Lalitha VenkataramananDual Neural Network Architecture for Determining Epistemic and Aleatoric Uncertainties
Chunlin Ji and Haige ShenStochastic Variational Inference via Upper Bound
Taylan Cemgil, Sumedh Ghaisas, Krishnamurthy Dvijotham and Pushmeet KohliLearning Perturbation-Invariant Representations with Smooth Encoders
Jack Fitzsimons, Sebastian Schmon and Stephen RobertsImplicit Priors for Knowledge Sharing in Bayesian Neural Networks
Mike Wu,Kristy Choi,Noah GoodmanandStefano ErmonMeta-Amortized Variational Inference and Learning
He Zhao, Piyush Rai, Lan Du, Wray Buntine, Dinh Phung and Mingyuan ZhouA Bayesian Extension to VAEs for Discrete Data
Agustinus Kristiadi, Sina Däubener and Asja FischerUncertainty quantification with compound density networks
Niklas Heim, Václav Šmídl and Tomáš PevnýRodent: Relevance determination in ODE
Kathrin Grosse, David Pfaff, Michael T. Smith and Michael BackesThe Limitations of Model Uncertainty in Adversarial Settings
Timon Willi, Jonathan Masci, Jürgen Schmidhuber and Christian OsendorferRecurrent Neural Processes
Samuel Kessler, Vu Nguyen, Stefan Zohren and Steve RobertsIndian Buffet Neural Networks for Continual Learning
Mariana Vargas Vieyra, Aurélien Bellet and Pascal DenisProbabilistic End-to-End Graph-based Semi-Supervised Learning
Apratim Bhattacharyya,Mario FritzandBernt Schiele“Best of Many” Samples Distribution Matching
Apratim Bhattacharyya, Michael Hanselmann,Mario Fritz,Bernt Schieleand Christoph-Nikolas StraehleConditional Flow Variational Autoencoders for Structured Sequence Prediction
Rahul Sharma, Abhishek Kumar and Piyush RaiRefined $\alpha$-Divergence Variational Inference via Rejection Sampling
Luis A. Perez Rey, Vlado Menkovski and Jacobus W. PortegiesCan VAEs capture topological properties?
Geoffroy Dubourg-Felonneau, Omar Darwish, Christopher Parsons, Dàmi Rebergen, John Cassidy, Nirmesh Patel and Harry CliffordDeep Bayesian Recurrent Neural Networks for Somatic Variant Calling in Cancer
Jonathan Warrell and Mark GersteinHierarchical PAC-Bayes Bounds via Deep Probabilistic Programming
Slava Voloshynovskiy,Mouad Kondah,Shideh Rezaeifar,Olga Taran,Taras HolotyakandDanilo J. RezendeInformation bottleneck through variational glasses
Mihaela Rosca, Michael Figurnov, Shakir Mohamed and Andriy MnihMeasure Valued Derivatives for Approximate Bayesian Inference
Max-Heinrich Laves, Sontje Ihler, Karl-Philipp Kortmann and Tobias OrtmaierWell-calibrated Model Uncertainty with Temperature Scaling for Dropout Variational Inference
Matthew Willetts, Stephen Roberts and Chris HolmesDisentangling to Cluster: Gaussian Mixture Variational Ladder Autoencoders
Jiaming Song and Stefano ErmonMutual Information Estimation as Optimization over Density Ratios: A Unifying Perspective
Andrew Ross, Jianzhun Du, Yonadav Shavit and Finale Doshi-VelezControlled Direct Effect Priors for Bayesian Neural Networks
Yeming Wen,Dustin TranandJimmy BaBatchEnsemble: Efficient Ensemble of Deep Neural Networks via Rank-1 Perturbation
Homa Fashandi and Darin GrahamEmpirical Studies on Sensitivity to Perturbations and Hyperparameters in Bayesian Neural Networks
Ruiqi Gao,Erik Nijkamp, Zhen Xu,Andrew Dai, Diederik Kingma andYing Nian WuFlow Contrastive Estimation of Energy-Based Model
Hooshmand Shokri Razaghi and Liam PaninskiFiltering Normalizing Flows
Joshua Chang, Shashaank Vattikuti and Carson ChowProbabilistically-autoencoded horseshoe-disentangled multidomain item-response theory models
Tim Xiao,Aidan GomezandYarin GalWat heb je gezegd? Detecting Out-of-Distribution Translations with Variational Transformers
Vanessa Boehm, Francois Lanusse and Uros SeljakUncertainty Quantification with Generative Models
Giorgio Giannone, Christian Osendorfer and Jonathan MasciNo Representation without Transformation
Gintare Karolina Roy, Waseem Gharbieh, Kyle Hsu andDaniel RoyOptimal (PAC-Bayes) priors are data dependent
Ranganath Krishnan, Mahesh Subedar, Omesh Tickoo, Angelos Filos and Yarin GalImproving MFVI in Bayesian Neural Networks with Empirical Bayes: a Study with Diabetic Retinopathy Diagnosis
Ari Heljakka, Yuxin Hou,Juho KannalaandArno SolinConditional Image Sampling by Deep Automodulators
Joe Davison, Kristen Severson and Soumya GhoshCross-population Variational Autoencoders
Arsenii Ashukha, Alexander Lyzhov,Dmitry MolchanovandDmitry VetrovPitfalls of In-Domain Uncertainty Estimation and Ensembling in Deep Learning
Jeremiah LiuVariable Selection with Rigorous Uncertainty Quantification using Bayesian Deep Neural Networks
Sicong Huang,Alireza Makhzani,Yanshuai CaoandRoger GrosseEvaluating Lossy Compression Rates of Deep Generative Models
Dian Ang Yap, Nicholas Roberts and Vinay PrabhuGrassmannian Packings in Neural Networks: Learning with Maximal Subspace Packings for Diversity and Anti-Sparsity
Evgenii Nikishin,Arsenii AshukhaandDmitry VetrovUnsupervised Domain Adaptation with Shared Latent Dynamics for Reinforcement Learning
Imant Daunhawer, Thomas Sutter and Julia E. VogtImproving Multimodal Generative Models with Disentangled Latent Partitions
Sunho Park, Saehoon Kim, Hongming Xu and Tae Hyun HwangDeep Gaussian processes for weakly supervised learning: tumor mutation burden (TMB) prediction
Nilesh Ahuja, Ibrahima Ndiour, Trushant Kalyanpur and Omesh TickooProbabilistic Modeling of Deep Features for Out-of-Distribution and Adversarial Detection
Stanislav Fort, Huiyi Hu andBalaji LakshminarayananDeep Ensembles: A Loss Landscape Perspective
Abhishek Kumar and Ben PooleOn Implicit Regularization in β-VAE
Steindor Saemundsson, Katja Hofmann and Marc DeisenrothVariational Integrator Networks
Mahesh Subedar, Nilesh Ahuja, Ranganath Krishnan, Ibrahima Ndiour and Omesh TickooDeep Probabilistic Models to Detect Data Poisoning Attacks
Roman Novak, Lechao Xiao, Jiri Hron, Jaehoon Lee, Jascha Sohl-Dickstein and Samuel SchoenholzNeural Tangents: Easy and Fast Infinite Networks in Python
Artyom Gadetsky, Kirill Struminsky, Christopher Robinson, Novi Quadrianto and Dmitry VetrovLow-variance Gradient Estimates for the Plackett-Luce Distribution
Jhosimar Arias FigueroaSemi-supervised Learning using Deep Generative Models and Auxiliary Tasks
Erik Daxberger and José Miguel Hernández-LobatoBayesian VAEs for Unsupervised Anomaly Detection
Xavier Gitiaux, Shane Maloney, Anna Jungbluth, Carl Shneider, Atılım Güneş Baydin, Paul J. Wright, Yarin Gal, Michel Deudon, Alfredo Kalaitzis and Andres Munoz-JaramilloProbabilistic Super-Resolution of Solar Magnetograms: Generating Many Explanations and Measuring Uncertainties
Tim G. J. Rudner, Florian Wenzel and Yarin GalThe Natural Neural Tangent Kernel: Neural Network Training Dynamics under Natural Gradient Descent
Waseem Aslam, Tim G. J. Rudner and Yarin GalTighter Variational Bounds for Deep Gaussian Processes
Adrián Csiszárik, Beatrix Benkő and Dániel VargaNegative Sampling in Variational Autoencoders
Aditya Grover, Dustin Tran, Rui Shu, Ben Poole and Kevin MurphyProbing Uncertainty Estimates of Neural Processes

Call for papers

We invite researchers to submit work in any of the following areas:

  • Uncertainty in deep learning,
  • probabilistic deep models (such as extensions and application of Bayesian neural networks),
  • deep probabilistic models (such as hierarchical Bayesian models and their applications),
  • deep generative models (such as variational autoencoders),
  • practical approximate inference techniques in Bayesian deep learning,
  • connections between deep learning and Gaussian processes,
  • applications of Bayesian deep learning,
  • or any of the topics below.

A submission should take the form of an extended abstract (3 pages long) in PDF format using the NeurIPS 2019 style. Author names do not need to be anonymized, and conflicts of interest in assessing submitted contributions will be based on this (reviewers will not be involved in the assessment of a submission by authors within the same institution). References may extend as far as needed beyond the 3 page upper limit. Submissions may extend beyond the 3 pages upper limit, but reviewers are not expected to read beyond the first 3 pages. If the research has previously appeared in a journal, workshop, or conference (including the NeurIPS 2019 conference), the workshop submission should extend that previous work. Dual submissions to ICLR 2019, AAAI 2019, and AISTATS 2019 are permitted.

Submissions will be accepted as contributed talks or poster presentations. Extended abstracts should be submitted by September 9, 2019 Deadline has been extended to Friday, September 13, 2019; submission page is here. Final versions will be posted on the workshop website (and are archival but do not constitute a proceedings). Notification of acceptance will be made before October 1, 2019.

Key Dates:

  • Extended abstract submission deadline: September 9, 2019 (23:59 AOE) Deadline has been extended to Friday, September 13, 2019 (23:59 AOE) (submission page is here)
  • Acceptance notification: before October 1, 2019
  • Camera ready submission: December 1, 2019
  • Workshop: Friday, December 13, 2019

Please make sure to apply to the lottery registration as the workshop is allocated a limited number of tickets. We will do our best to guarantee workshop registration for all accepted workshop submissions which did not receive a lottery registration.


While deep learning has been revolutionary for machine learning, most modern deep learning models cannot represent their uncertainty nor take advantage of the well studied tools of probability theory. This has started to change following recent developments of tools and techniques combining Bayesian approaches with deep learning. The intersection of the two fields has received great interest from the community over the past few years, with the introduction of new deep learning models that take advantage of Bayesian techniques, as well as Bayesian models that incorporate deep learning elements [1-11]. In fact, the use of Bayesian techniques in deep learning can be traced back to the 1990s’, in seminal works by Radford Neal [12], David MacKay [13], and Dayan et al. [14]. These gave us tools to reason about deep models’ confidence, and achieved state-of-the-art performance on many tasks. However earlier tools did not adapt when new needs arose (such as scalability to big data), and were consequently forgotten. Such ideas are now being revisited in light of new advances in the field, yielding many exciting new results.

Extending on last year’s workshop’s success, this workshop will again study the advantages and disadvantages of such ideas, and will be a platform to host the recent flourish of ideas using Bayesian approaches in deep learning and using deep learning tools in Bayesian modelling. The program includes a mix of invited talks, contributed talks, and contributed posters. It will be composed of five themes: deep generative models, variational inference using neural network recognition models, practical approximate inference techniques in Bayesian neural networks, applications of Bayesian neural networks, and information theory in deep learning. Future directions for the field will be debated in a panel discussion.

Previous workshops:

Our 2018 workshop page is available here; Our 2017 workshop page is available here; Our 2016 workshop page is available here; videos from the 2016 workshop are available online as well.


  • Uncertainty in deep learning,
  • Applications of Bayesian deep learning,
  • Probabilistic deep models (such as extensions and application of Bayesian neural networks),
  • Deep probabilistic models (such as hierarchical Bayesian models and their applications),
  • Generative deep models (such as variational autoencoders),
  • Information theory in deep learning,
  • Deep ensemble uncertainty,
  • NTK and Bayesian modelling,
  • Connections between NNs and GPs,
  • Incorporating explicit prior knowledge in deep learning (such as posterior regularisation with logic rules),
  • Approximate inference for Bayesian deep learning (such as variational Bayes / expectation propagation / etc. in Bayesian neural networks),
  • Scalable MCMC inference in Bayesian deep models,
  • Deep recognition models for variational inference (amortised inference),
  • Bayesian deep reinforcement learning,
  • Deep learning with small data,
  • Deep learning in Bayesian modelling,
  • Probabilistic semi-supervised learning techniques,
  • Active learning and Bayesian optimisation for experimental design,
  • Kernel methods in Bayesian deep learning,
  • Implicit inference,
  • Applying non-parametric methods, one-shot learning, and Bayesian deep learning in general.


  1. Kingma, DP and Welling, M, ‘’Auto-encoding variational bayes’’, 2013.
  2. Rezende, D, Mohamed, S, and Wierstra, D, ‘’Stochastic backpropagation and approximate inference in deep generative models’’, 2014.
  3. Blundell, C, Cornebise, J, Kavukcuoglu, K, and Wierstra, D, ‘’Weight uncertainty in neural network’’, 2015.
  4. Hernandez-Lobato, JM and Adams, R, ’’Probabilistic backpropagation for scalable learning of Bayesian neural networks’’, 2015.
  5. Gal, Y and Ghahramani, Z, ‘’Dropout as a Bayesian approximation: Representing model uncertainty in deep learning’’, 2015.
  6. Gal, Y and Ghahramani, G, ‘’Bayesian convolutional neural networks with Bernoulli approximate variational inference’’, 2015.
  7. Kingma, D, Salimans, T, and Welling, M. ‘’Variational dropout and the local reparameterization trick’’, 2015.
  8. Balan, AK, Rathod, V, Murphy, KP, and Welling, M, ‘’Bayesian dark knowledge’’, 2015.
  9. Louizos, C and Welling, M, “Structured and Efficient Variational Deep Learning with Matrix Gaussian Posteriors”, 2016.
  10. Lawrence, ND and Quinonero-Candela, J, “Local distance preservation in the GP-LVM through back constraints”, 2006.
  11. Tran, D, Ranganath, R, and Blei, DM, “Variational Gaussian Process”, 2015.
  12. Neal, R, ‘’Bayesian Learning for Neural Networks’’, 1996.
  13. MacKay, D, ‘’A practical Bayesian framework for backpropagation networks‘’, 1992.
  14. Dayan, P, Hinton, G, Neal, R, and Zemel, S, ‘’The Helmholtz machine’’, 1995.
  15. Wilson, AG, Hu, Z, Salakhutdinov, R, and Xing, EP, “Deep Kernel Learning”, 2016.
  16. Saatchi, Y and Wilson, AG, “Bayesian GAN”, 2017.
  17. MacKay, D.J.C. “Bayesian Methods for Adaptive Models”, PhD thesis, 1992.