mlsanityData to 14 Oct 2026

Low-rank extended Kalman filtering for online learning of neural networks from streaming data

arXiv 2305.19535 ↗First posted 31 May 2023stat.ML cs.LGcited by 23

Not on the map. Why?

Abstract (arXiv)

We propose an efficient online approximate Bayesian inference algorithm for estimating the parameters of a nonlinear function from a potentially non-stationary data stream. The method is based on the extended Kalman filter (EKF), but uses a novel low-rank plus diagonal decomposition of the posterior precision matrix, which gives a cost per step which is linear in the number of model parameters. In contrast to methods based on stochastic variational inference, our method is fully deterministic, and does not require step-size tuning. We show experimentally that this results in much faster (more sample efficient) learning, which results in more rapid adaptation to changing distributions, and faster accumulation of reward when used as part of a contextual bandit algorithm.

Key facts. Cited by 23 papers in the map (data through 14 Oct 2026). Builds on 42 of its 83 references that are in the map. First posted on arXiv 31 May 2023.

Builds on

42 of 83 references.

References · 42

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  3. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  4. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,349
  5. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
  6. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 872
  7. 2015Optimizing Neural Networks with Kronecker-factored Approximate Curvaturecited by 796
  8. 2022Continual Test-Time Domain Adaptationcited by 525
  9. 2017A Tutorial on Thompson Samplingcited by 433
  10. 2021Laplace Redux -- Effortless Bayesian Deep Learningcited by 365
  11. 2017Variational Continual Learningcited by 339
  12. 2015Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networkscited by 295
30 more
  1. 2020Towards Continual Reinforcement Learning: A Review and Perspectivescited by 274
  2. 2021Online Continual Learning in Image Classification: An Empirical Surveycited by 245
  3. 2018Re-evaluating Continual Learning Scenarios: A Categorization and Case for Strong Baselinescited by 228
  4. 2018Online Structured Laplace Approximations For Overcoming Catastrophic Forgettingcited by 201
  5. 2020Continual Prototype Evolution: Learning Online from Non-Stationary Data Streamscited by 144
  6. 2021Continual World: A Robotic Benchmark For Continual Reinforcement Learningcited by 116
  7. 2019Continual Learning for Robotics: Definition, Framework, Learning Strategies, Opportunities and Challengescited by 93
  8. 2020A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learningcited by 87
  9. 2021Continual Backprop: Stochastic Gradient Descent with Persistent Randomnesscited by 80
  10. 2022A Framework and Benchmark for Deep Batch Active Learning for Regressioncited by 49
  11. 2013Streaming Variational Bayescited by 48
  12. 2018SLANG: Fast Structured Covariance Approximations for Bayesian Deep Learning with Natural Gradientcited by 34
  13. 2020Task Agnostic Continual Learning Using Online Variational Bayes with Fixed-Point Updatescited by 34
  14. 2020Fast Predictive Uncertainty for Classification with Bayesian Deep Networkscited by 31
  15. 2017Online Natural Gradient as a Kalman Filtercited by 25
  16. 2020Lifelong Incremental Reinforcement Learning with Online Bayesian Inferencecited by 23
  17. 2017Gaussian variational approximation with a factor covariance structurecited by 22
  18. 2020Cold Posteriors and Aleatoric Uncertaintycited by 22
  19. 2017Semi-analytical approximations to statistical moments of sigmoid and softmax mappings of normal variablescited by 20
  20. 2021Knowledge-Adaptation Priorscited by 20
  21. 2019Thompson Sampling with Approximate Inferencecited by 15
  22. 2021Efficient Online Bayesian Inference for Neural Banditscited by 15
  23. 2022ORFit: One-Pass Learning via Bridging Orthogonal Gradient Descent and Recursive Least-Squarescited by 12
  24. 2021Kalman Bayesian Neural Networks for Closed-form Online Learningcited by 9
  25. 2023Particle-based Online Bayesian Samplingcited by 7
  26. 2023The limited-memory recursive variational Gaussian approximation (L-RVGA)cited by 6
  27. 2018Efficient Full-Matrix Adaptive Regularizationcited by 5
  28. 2021Viking: Variational Bayesian Variance Trackingcited by 4
  29. 2021The Fragility of Noise Estimation in Kalman Filter: Optimization Can Handle Model-Misspecificationcited by 3
  30. 2018Deterministic Variational Inference for Robust Bayesian Neural Networkscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 2; 2024: 11; 2025: 5; 2026: 5.

23 papers.

Sort:

By area:

  1. 2025On Vanishing Gradients, Over-Smoothing, and Over-Squashing in GNNs: Bridging Recurrent and Graph Learningcited by 43
  2. 2024Outlier-robust Kalman Filtering through Generalised Bayescited by 21
  3. 2024Non-Stationary Learning of Neural Networks with Automatic Soft Parameter Resetcited by 17
  4. 2024Rough Transformers: Lightweight and Continuous Time Series Modelling through Signature Patchingcited by 16
  5. 2023Kalman Filter for Online Classification of Non-Stationary Datacited by 15
  6. 2024Bayesian Online Natural Gradient (BONG)cited by 14
  7. 2024Temporal Test-Time Adaptation with State-Space Modelscited by 7
  8. 2024A unifying framework for generalised Bayesian online learning in non-stationary environmentscited by 5
  9. 2023Implicit Maximum a Posteriori Filtering via Adaptive Optimizationcited by 4
  10. 2024Stragglers-Aware Low-Latency Synchronous Federated Learning via Layer-Wise Model Updatescited by 4
  11. 2025Martingale Posterior Neural Networks for Fast Sequential Decision Makingcited by 4
  12. 2024Numerically Robust Fixed-Point Smoothing Without State Augmentationcited by 2
11 more
  1. 2025KOALA++: Efficient Kalman-Based Optimization with Gradient-Covariance Productscited by 2
  2. 2024Adjusting Model Size in Continual Gaussian Processes: How Big is Big Enough?cited by 1
  3. 2024LoKO: Low-Rank Kalman Optimizer for Online Fine-Tuning of Large Modelscited by 1
  4. 2025Bayesian Natural Gradient Fine-Tuning of CLIP Models via Kalman Filteringcited by 1
  5. 2026Gradient Regularized Natural Gradientscited by 1
  6. 2026The Dynamic-Probabilistic Consistency Gap in Chaotic Surrogate Modelingcited by 1
  7. 2024Noise misleads rotation invariant algorithms on sparse targetscited by 0
  8. 2025Staying Alive: Online Neural Network Maintenance and Systemic Driftcited by 0
  9. 2026Using the SEKF to Transfer NN Models of Dynamical Systems with Limited Datacited by 0
  10. 2026Infinite-Parameter LLMs: Generating and Adapting Weights from Live Datacited by 0
  11. 2026Online Learning via Learned Latent Bayesian Trackingcited by 0