mlsanityData to 7 Oct 2026

Flow-Induced Diagonal Gaussian Processes

arXiv 2509.17153 ↗First posted 21 Sep 2025cs.LG cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

We present Flow-Induced Diagonal Gaussian Processes (FiD-GP), a compression framework that incorporates a compact inducing weight matrix to project a neural network's weight uncertainty into a lower-dimensional subspace. Critically, FiD-GP relies on normalising-flow priors and spectral regularisations to augment its expressiveness and align the inducing subspace with feature-gradient geometry through a numerically stable projection mechanism objective. Furthermore, we demonstrate how the prediction framework in FiD-GP can help to design a single-pass projection for Out-of-Distribution (OoD) detection. Our analysis shows that FiD-GP improves uncertainty estimation ability on various tasks compared with SVGP-based baselines, satisfies tight spectral residual bounds with theoretically guaranteed OoD detection, and significantly compresses the neural network's storage requirements at the cost of increased inference computation dependent on the number of inducing weights employed. Specifically, in a comprehensive empirical study spanning regression, image classification, semantic segmentation, and out-of-distribution detection benchmarks, it cuts Bayesian training cost by several orders of magnitude, compresses parameters by roughly 51%, reduces model size by about 75%, and matches state-of-the-art accuracy and uncertainty estimation.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 32 of its 62 references that are in the map. First posted on arXiv 21 Sep 2025.

Builds on

32 of 62 references.

References · 32

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  3. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  4. 2015Variational Inference with Normalizing Flowscited by 2,161
  5. 2016Density estimation using Real NVPcited by 2,029
  6. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  7. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,539
  8. 2016Variational Inference: A Review for Statisticianscited by 1,466
  9. 2017Masked Autoregressive Flow for Density Estimationcited by 680
  10. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 636
  11. 2018FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Modelscited by 584
  12. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 560
20 more
  1. 2019Likelihood Ratios for Out-of-Distribution Detectioncited by 451
  2. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  3. 2020Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awarenesscited by 389
  4. 2015Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networkscited by 295
  5. 2019High-Resolution Representations for Labeling Pixels and Regionscited by 211
  6. 2021Deep Deterministic Uncertainty: A Simple Baselinecited by 200
  7. 2020Efficient and Scalable Bayesian Neural Nets with Rank-1 Factorscited by 171
  8. 2020Uncertainty Quantification and Deep Ensemblescited by 138
  9. 2020A Survey of Constrained Gaussian Process Regression: Approaches and Implementation Challengescited by 51
  10. 2015MCMC for Variationally Sparse Gaussian Processescited by 44
  11. 2017Convolutional Gaussian Processescited by 44
  12. 2018SLANG: Fast Structured Covariance Approximations for Bayesian Deep Learning with Natural Gradientcited by 34
  13. 2016Preconditioning Kernel Matricescited by 32
  14. 2020A Tutorial on Sparse Gaussian Processes and Variational Inferencecited by 29
  15. 2020Encoding the latent posterior of Bayesian Neural Networks for uncertainty quantificationcited by 17
  16. 2021Sparse Uncertainty Representation in Deep Learning with Inducing Weightscited by 16
  17. 2021Uncertainty Quantification by Ensemble Learning for Computational Optical Form Measurementscited by 11
  18. 2024Scalable Bayesian Optimization via Focalized Sparse Gaussian Processescited by 8
  19. 2024Uncertainty-Aware Out-of-Distribution Detection with Gaussian Processescited by 6
  20. 2024Discriminant Distance-Aware Representation on Deterministic Uncertainty Quantification Methodscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Bayesian-LoRA: Probabilistic Low-Rank Adaptation of Large Language Modelscited by 2