An Updating Survey for Bayesian Deep Learning (BDL)

This is an updating survey for Bayesian Deep Learning (BDL), an constantly updated and extended version for the manuscript, 'A Survey on Bayesian Deep Learning', published in ACM Computing Surveys 2020.<br>

Bayesian deep learning is a powerful framework for designing models across a wide range of applications. See our Nature Medicine paper for a possible application on healthcare.

Contents

• Survey
• BDL and Recommender Systems
• BDL and Domain Adaptation (and Domain Generalization, Meta Learning, etc.)
• BDL and Healthcare
• BDL and Natural Language Processing (NLP)
• BDL and Computer Vision (CV)
• BDL and Control/Planning
• BDL and Graphs (Link Prediction, Graph Neural Networks, Knowledge Graphs, etc.)
• BDL and Topic Modeling
• BDL and Speech Recognition/Synthesis
• BDL and Forecasting (Time Series Analysis)
• BDL and Distributed/Federated Learning
• BDL and Continual/Life-Long Learning
• BDL and AI4Science
• BDL as a Framework (Miscellaneous)
• Bayesian/Probabilistic Neural Networks as Building Blocks of BDL

Survey

A Survey on Bayesian Deep Learning<br> by Wang et al., ACM Computing Surveys (CSUR) 2020<br> [PDF] [Blog] [BDL Framework in 2016]

<p align="center"> <img src="./BDL_Table.png" alt="" data-canonical-src="./BDL_Table.png" width="930" height="580"/> </p>

BDL and Recommender Systems

Collaborative Deep Learning for Recommender Systems<br> by Wang et al., KDD 2015<br> [PDF] [Project Page] [2014 Arxiv Version] [Code] [MXNet Code] [TensorFlow Code] [Dataset A] [Dataset B] [Jupyter Notebook] [Slides] [Slides (Long)]

Collaborative Recurrent Autoencoder: Recommend while Learning to Fill in the Blanks<br> by Wang et al., NIPS 2016<br> [PDF]

Collaborative Knowledge Base Embedding for Recommender Systems<br> by Zhang et al., KDD 2016<br> [PDF]

Collaborative Deep Ranking: A Hybrid Pair-Wise Recommendation Algorithm with Implicit Feedback<br> by Ying et al., PAKDD 2016<br> [PDF]

Collaborative Variational Autoencoder for Recommender Systems<br> by Li et al., KDD 2017<br> [PDF]

Variational Autoencoders for Collaborative Filtering<br> by Liang et al., WWW 2018<br> [PDF]

Probabilistic Metric Learning with Adaptive Margin for Top-K Recommendation<br> by Ma et al., KDD 2020<br> [PDF]

BDL and Domain Adaptation (and Domain Generalization, Meta Learning, etc.)

Probabilistic Model-Agnostic Meta-Learning<br> by Finn et al., NIPS 2018<br> [PDF]

Bayesian Model-Agnostic Meta-Learning<br> by Yoon et al., NIPS 2018<br> [PDF]

Recasting Gradient-Based Meta-Learning as Hierarchical Bayes<br> by Grant et al., ICLR 2018<br> [PDF]

Reconciling Meta-Learning and Continual Learning with Online Mixtures of Tasks<br> by Jerfal et al., NIPS 2019<br> [PDF]

Meta-Learning Probabilistic Inference For Prediction<br> by Gordon et al., ICLR 2019<br> [PDF]

Learning to Learn with Variational Information Bottleneck for Domain Generalization<br> by Du et al., ECCV 2020<br> [PDF]

Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels<br> by Patacchiola et al., NIPS 2020<br> [PDF]

Continuously Indexed Domain Adaptation<br> by Wang et al., ICML 2020<br> [PDF]

A Bit More Bayesian: Domain-Invariant Learning with Uncertainty<br> by Xiao et al., ICML 2021<br> [PDF]

Domain-Indexing Variational Bayes: Interpretable Domain Index for Domain Adaptation<br> by Xu et al., ICLR 2023<br> [PDF]

BDL and Healthcare

Electronic Health Record Analysis via Deep Poisson Factor Models<br> by Henao et al., JMLR 2016<br> [PDF]

Structured Inference Networks for Nonlinear State Space Models<br> by Krishnan et al., AAAI 2017<br> [PDF]

Causal Effect Inference with Deep Latent-Variable Models<br> by Louizos et al., NIPS 2017<br> [PDF]

Black Box FDR<br> by Tansey et al., ICML 2018<br> [PDF]

Bidirectional Inference Networks: A Class of Deep Bayesian Networks for Health Profiling<br> by Wang et al., AAAI 2019<br> [PDF]

Sampling-free Uncertainty Estimation in Gated Recurrent Units with Applications to Normative Modeling in Neuroimaging<br> by Hwang et al., UAI 2019<br> [PDF]

Neural Jump Stochastic Differential Equations<br> by Jia et al., NIPS 2019<br> [PDF]

Towards Interpretable Clinical Diagnosis with Bayesian Network Ensembles Stacked on Entity-Aware CNNs<br> by Chen et al., ACL 2020<br> [PDF]

Continuously Indexed Domain Adaptation<br> by Wang et al., ICML 2020<br> [PDF] [Cross Referenced in BDL and Domain Adaptation]

Assessment of medication self-administration using artificial intelligence<br> by Zhao et al., Nature Medicine 2021<br> [PDF]

Neural Pharmacodynamic State Space Modeling<br> by Hussain et al., ICML 2021<br> [PDF]

Self-Interpretable Time Series Prediction with Counterfactual Explanations<br> by Yan et al., ICML 2023<br> [PDF] [Cross Referenced in BDL and Forecasting (Time Series Analysis)]

BDL and NLP

Sequence to Better Sequence: Continuous Revision of Combinatorial Structures<br> by Mueller et al., ICML 2017<br> [PDF]

QuaSE: Sequence Editing under Quantifiable Guidance<br> by Liao et al., EMNLP 2018<br> [PDF]

Dispersed Exponential Family Mixture VAEs for Interpretable Text Generation<br> by Shi et al., ICML 2020<br> [PDF]

Towards Interpretable Clinical Diagnosis with Bayesian Network Ensembles Stacked on Entity-Aware CNNs<br> by Chen et al., ACL 2020<br> [PDF] [Cross Referenced in BDL and Healthcare]

What You Say and How You Say it: Joint Modeling of Topics and Discourse in Microblog Conversations<br> by Zeng et al., ACL 2020<br> [PDF]

Latent Diffusion Energy-Based Model for Interpretable Text Modeling<br> by Yu et al., ICML 2022<br> [PDF]

Diffusion-LM Improves Controllable Text Generation<br> by Li et al., NeurIPS 2022<br> [PDF]

Tractable Control for Autoregressive Language Generation<br> by Zhang et al., ICML 2023<br> [PDF]

BDL and Computer Vision

Attend, Infer, Repeat: Fast Scene Understanding with Generative Models<br> by Eslami et al., NIPS 2016<br> [PDF]

Efficient Inference in Occlusion-aware Generative Models of Images<br> by Huang et al., ICLR 2016<br> [PDF]

Sequential Attend, Infer, Repeat: Generative Modelling of Moving Objects<br> by Kosiorek et al., NIPS 2018<br> [PDF]

Gaussian Process Prior Variational Autoencoders<br> by Casale et al., NIPS 2018<br> [PDF]

Spatially Invariant Unsupervised Object Detection with Convolutional Neural Networks<br> by Crawford et al., AAAI 2019<br> [PDF]

Faster Attend-Infer-Repeat with Tractable Probabilistic Models<br> by Stelzner et al., ICML 2019<br> [PDF]

Asynchronous Temporal Fields for Action Recognition<br> by Sigurdsson et al., CVPR 2017<br> [PDF]

Generalizing Eye Tracking with Bayesian Adversarial Learning<br> by Wang et al., CVPR 2019<br> [PDF]

Sequential Neural Processes<br> by Singh et al., NIPS 2019<br> [PDF]

SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decomposition<br> by Lin et al., ICLR 2020<br> [PDF]

Being Bayesian about Categorical Probability<br> by Joo et al., ICML 2020<br> [PDF]

NVAE: A Deep Hierarchical Variational Autoencoder<br> by Vahdat et al., NIPS 2020<br> [PDF]

Learning Latent Space Energy-Based Prior Model<br> by Pang et al., NIPS 2020<br> [PDF]

Generative Neurosymbolic Machines<br> by Jiang et al., NIPS 2020<br> [PDF]

Denoising Diffusion Probabilistic Models<br> by Ho et al., NIPS 2020<br> [PDF]

A Causal View of Compositional Zero-shot Recognition<br> by Atzmon et al., NIPS 2020<br> [PDF]

Counterfactuals Uncover the Modular Structure of Deep Generative Models<br> by Besserve et al., ICLR 2020<br> [PDF]

ROOTS: Object-Centric Representation and Rendering of 3D Scenes<br> by Chen et al., JMLR 2021<br> [PDF]