mlsanityData to 7 Oct 2026

A Comprehensive guide to Bayesian Convolutional Neural Network with Variational Inference

arXiv 1901.02731 ↗First posted 8 Jan 2019cs.LG stat.MLcited by 51

Not on the map. Why?

Abstract (arXiv)

Artificial Neural Networks are connectionist systems that perform a given task by learning on examples without having prior knowledge about the task. This is done by finding an optimal point estimate for the weights in every node. Generally, the network using point estimates as weights perform well with large datasets, but they fail to express uncertainty in regions with little or no data, leading to overconfident decisions. In this paper, Bayesian Convolutional Neural Network (BayesCNN) using Variational Inference is proposed, that introduces probability distribution over the weights. Furthermore, the proposed BayesCNN architecture is applied to tasks like Image Classification, Image Super-Resolution and Generative Adversarial Networks. The results are compared to point-estimates based architectures on MNIST, CIFAR-10 and CIFAR-100 datasets for Image CLassification task, on BSD300 dataset for Image Super Resolution task and on CIFAR10 dataset again for Generative Adversarial Network task. BayesCNN is based on Bayes by Backprop which derives a variational approximation to the true posterior. We, therefore, introduce the idea of applying two convolutional operations, one for the mean and one for the variance. Our proposed method not only achieves performances equivalent to frequentist inference in identical architectures but also incorporate a measurement for uncertainties and regularisation. It further eliminates the use of dropout in the model. Moreover, we predict how certain the model prediction is based on the epistemic and aleatoric uncertainties and empirically show how the uncertainty can decrease, allowing the decisions made by the network to become more deterministic as the training accuracy increases. Finally, we propose ways to prune the Bayesian architecture and to make it more computational and time effective.

Key facts. Cited by 51 papers in the map (data through 7 Oct 2026). First posted on arXiv 8 Jan 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 12; 2022: 8; 2023: 6; 2024: 8; 2025: 10; 2026: 7.

51 papers.

Sort:

By area:

  1. 2023Bayesian Low-rank Adaptation for Large Language Modelscited by 112
  2. 2022Trustworthy clinical AI solutions: a unified review of uncertainty quantification in deep learning models for medical image analysiscited by 62
  3. 2024A Survey of Uncertainty Estimation in LLMs: Theory Meets Practicecited by 54
  4. 2023Uncertainty-based Fairness Measurescited by 15
  5. 2025Grammars of Formal Uncertainty: When to Trust LLMs in Automated Reasoning Taskscited by 9
  6. 2023Bayesian neural networks via MCMC: a Python-based tutorialcited by 8
  7. 2022Reducing Overconfidence Predictions for Autonomous Driving Perceptioncited by 7
  8. 2021Differentially Private Bayesian Neural Networks on Accuracy, Privacy and Reliabilitycited by 6
  9. 2021Modeling Annotation Uncertainty with Gaussian Heatmaps in Landmark Localizationcited by 6
  10. 2022Wavelength-aware 2D Convolutions for Hyperspectral Imagingcited by 6
  11. 2024Bayesian neural networks for predicting uncertainty in full-field material responsecited by 5
  12. 2021An Uncertainty-Informed Framework for Trustworthy Fault Diagnosis in Safety-Critical Applicationscited by 4
38 more
  1. 2024X-Fake: Juggling Utility Evaluation and Explanation of Simulated SAR Imagescited by 4
  2. 2024TapType: Ten-finger text entry on everyday surfaces via Bayesian inferencecited by 4
  3. 2025Delusions of Large Language Modelscited by 4
  4. 2021Recurrent Localization Networks applied to the Lippmann-Schwinger Equationcited by 3
  5. 2021W2WNet: a two-module probabilistic Convolutional Neural Network with embedded data cleansing functionalitycited by 3
  6. 2021Shift-BNN: Highly-Efficient Probabilistic Bayesian Neural Network Training via Memory-Friendly Pattern Retrievingcited by 3
  7. 2021A Deterministic Sampling Method via Maximum Mean Discrepancy Flow with Adaptive Kernelcited by 3
  8. 2024EffLoc: Lightweight Vision Transformer for Efficient 6-DOF Camera Relocalizationcited by 3
  9. 2025Distance-informed Neural Processescited by 3
  10. 2021Multi Agent System for Machine Learning Under Uncertainty in Cyber Physical Manufacturing Systemcited by 2
  11. 2022Bayesian Convolutional Neural Networks for Limited Data Hyperspectral Remote Sensing Image Classificationcited by 2
  12. 2023Uncertainty-aware U-Net for Medical Landmark Detectioncited by 2
  13. 2021A Survey and Analysis on Automated Glioma Brain Tumor Segmentation and Overall Patient Survival Predictioncited by 1
  14. 2021A Bayesian Approach to Invariant Deep Neural Networkscited by 1
  15. 2022Cluster-based Regression using Variational Inference and Applications in Financial Forecastingcited by 1
  16. 2022Latent Preserving Generative Adversarial Network for Imbalance classificationcited by 1
  17. 2023SACDNet: Towards Early Type 2 Diabetes Prediction with Uncertainty for Electronic Health Recordscited by 1
  18. 2024Hierarchical Prompts for Rehearsal-free Continual Learningcited by 1
  19. 2024A Framework for Variational Inference of Lightweight Bayesian Neural Networks with Heteroscedastic Uncertaintiescited by 1
  20. 2025Accelerating Hamiltonian Monte Carlo for Bayesian Inference in Neural Networks and Neural Operatorscited by 1
  21. 2026Multi-Teacher Knowledge Distillation via Teacher-Informed Mixture Priorscited by 1
  22. 2021Scalable Bayesian Deep Learning with Kernel Seed Networkscited by 0
  23. 2021Improving the Behaviour of Vision Transformers with Token-consistent Stochastic Layerscited by 0
  24. 2022SoftDropConnect (SDC) -- Effective and Efficient Quantification of the Network Uncertainty in Deep MR Image Analysiscited by 0
  25. 2022Bayesian Layer Graph Convolutioanl Network for Hyperspetral Image Classificationcited by 0
  26. 2023Be Bayesian by Attachments to Catch More Uncertaintycited by 0
  27. 2024Revised Regularization for Efficient Continual Learning through Correlation-Based Parameter Update in Bayesian Neural Networkscited by 0
  28. 2025Logit Disagreement: OoD Detection with Bayesian Neural Networkscited by 0
  29. 2025Feature Preserving Shrinkage on Bayesian Neural Networks via the R2D2 Priorcited by 0
  30. 2025E-BayesSAM: Efficient Bayesian Adaptation of SAM with Self-Optimizing KAN-Based Interpretation for Uncertainty-Aware Ultrasonic Segmentationcited by 0
  31. 2025Ensemble Distribution Distillation for Self-Supervised Human Activity Recognitioncited by 0
  32. 2025Out-of-Distribution Detection from Small Training Sets using Bayesian Neural Network Classifierscited by 0
  33. 2025As If We've Met Before: LLMs Exhibit Certainty in Recognizing Seen Filescited by 0
  34. 2026Calibrated uncertainty quantification for prosumer flexibility aggregation in ancillary service marketscited by 0
  35. 2026Beyond Accuracy: Reliability and Uncertainty Estimation in Convolutional Neural Networkscited by 0
  36. 2026Probabilistic Federated Learning on Uncertain and Heterogeneous Data with Model Personalizationcited by 0
  37. 2026Evidence-Guided Neural Architecture Selection under Uncertainty for Subject-Specific Blood Glucose Forecastingcited by 0
  38. 2026Bayesian 3D Steerable CNNs: Enabling Equivariance and Uncertainty Quantification Simultaneouslycited by 0

50 most cited of 51 All 51 (JSON)