mlsanityData to 14 Oct 2026

Continual Learning via Sequential Function-Space Variational Inference

arXiv 2312.17210 ↗First posted 28 Dec 2023stat.ML cs.AI cs.LGcited by 29

Not on the map. Why?

Abstract (arXiv)

Sequential Bayesian inference over predictive functions is a natural framework for continual learning from streams of data. However, applying it to neural networks has proved challenging in practice. Addressing the drawbacks of existing techniques, we propose an optimization objective derived by formulating continual learning as sequential function-space variational inference. In contrast to existing methods that regularize neural network parameters directly, this objective allows parameters to vary widely during training, enabling better adaptation to new tasks. Compared to objectives that directly regularize neural network predictions, the proposed objective allows for more flexible variational distributions and more effective regularization. We demonstrate that, across a range of task sequences, neural networks trained via sequential function-space variational inference achieve better predictive accuracy than networks trained with related methods while depending less on maintaining a set of representative points from previous tasks.

Key facts. Cited by 29 papers in the map (data through 14 Oct 2026). Builds on 32 of its 47 references that are in the map. First posted on arXiv 28 Dec 2023.

Builds on

32 of 47 references.

References · 32

  1. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  2. 2016Learning without Forgettingcited by 2,496
  3. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  4. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
  5. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,105
  6. 2020Dark Experience for General Continual Learning: a Strong, Simple Baselinecited by 911
  7. 2018Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigencecited by 742
  8. 2018Progress & Compress: A scalable framework for continual learningcited by 520
  9. 2017Variational Continual Learningcited by 339
  10. 2017Overcoming Catastrophic Forgetting by Incremental Moment Matchingcited by 243
  11. 2018Online Structured Laplace Approximations For Overcoming Catastrophic Forgettingcited by 201
  12. 2018Towards Robust Evaluations of Continual Learningcited by 148
20 more
  1. 2019Uncertainty-based Continual Learning with Adaptive Regularizationcited by 146
  2. 2018Rotate your Networks: Better Weight Consolidation and Less Catastrophic Forgettingcited by 127
  3. 2019Functional Variational Bayesian Neural Networkscited by 127
  4. 2019Functional Regularisation for Continual Learning with Gaussian Processescited by 123
  5. 2020Continual Deep Learning by Functional Regularisation of Memorable Pastcited by 105
  6. 2019Uncertainty-guided Continual Learning with Bayesian Neural Networkscited by 102
  7. 2018Measuring and regularizing networks in function spacecited by 99
  8. 2015On Sparse variational methods and the Kullback-Leibler divergence between stochastic processescited by 57
  9. 2020Generalized Variational Continual Learningcited by 56
  10. 2013Streaming Variational Bayescited by 48
  11. 2017Streaming Sparse Gaussian Process Approximationscited by 41
  12. 2020Understanding Variational Inference in Function-Spacecited by 40
  13. 2017Less-forgetful Learning for Domain Expansion in Deep Neural Networkscited by 27
  14. 2020Variational Auto-Regressive Gaussian Processes for Continual Learningcited by 23
  15. 2019Continual Learning by Asymmetric Loss Approximation with Single-Side Overestimationcited by 22
  16. 2019Improving and Understanding Variational Continual Learningcited by 12
  17. 2019Hierarchical Indian Buffet Neural Networks for Bayesian Continual Learningcited by 11
  18. 2020Optimization and Generalization of Regularization-Based Continual Learning: a Loss Approximation Viewpointcited by 10
  19. 2018Keep and Learn: Continual Learning by Constraining the Latent Space for Knowledge Preservation in Neural Networkscited by 9
  20. 2019Continual Multi-task Gaussian Processescited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 12; 2025: 10; 2026: 7.

29 papers.

Sort:

By area:

  1. 2024Position: Bayesian Deep Learning is Needed in the Age of Large-Scale AIcited by 62
  2. 2024Make Continual Learning Stronger via C-Flatcited by 35
  3. 2024A Unified and General Framework for Continual Learningcited by 20
  4. 2024Function-space Parameterization of Neural Networks for Sequential Learningcited by 8
  5. 2024Continual Learning with Neuromorphic Computing: Foundations, Methods, and Emerging Applicationscited by 8
  6. 2025Dynamic Prototype Rehearsal for Continual ECG Arrhythmia Detectioncited by 4
  7. 2024Regularized KL-Divergence for Well-Defined Function-Space Variational Inference in Bayesian neural networkscited by 3
  8. 2024Parameter-Efficient Fine-Tuning for Continual Learning: A Neural Tangent Kernel Perspectivecited by 3
  9. 2025ErrorEraser: Unlearning Data Bias for Improved Continual Learningcited by 3
  10. 2024Attacking Bayes: On the Adversarial Robustness of Bayesian Neural Networkscited by 2
  11. 2024Adjusting Model Size in Continual Gaussian Processes: How Big is Big Enough?cited by 1
  12. 2024Temporal-Difference Variational Continual Learningcited by 1
17 more
  1. 2025Revisiting Clustering of Neural Bandits: Selective Reinitialization for Mitigating Loss of Plasticitycited by 1
  2. 2025C-Flat++: Towards a More Efficient and Powerful Framework for Continual Learningcited by 1
  3. 2025Understanding Catastrophic Interference: On the Identifibility of Latent Representationscited by 1
  4. 2026Exploring the Impact of Parameter Update Magnitude on Forgetting and Generalization of Continual Learningcited by 1
  5. 2026A Faster Path to Continual Learningcited by 1
  6. 2024On Sequential Maximum a Posteriori Inference for Continual Learningcited by 0
  7. 2024Balanced Gradient Sample Retrieval for Enhanced Knowledge Retention in Proxy-based Continual Learningcited by 0
  8. 2025Variational Bayesian Pseudo-Coresetcited by 0
  9. 2025Sequential Function-Space Variational Inference via Gaussian Mixture Approximationcited by 0
  10. 2025Continual learning via probabilistic exchangeable sequence modellingcited by 0
  11. 2025Prediction-Oriented Subsampling from Data Streamscited by 0
  12. 2025Monte Carlo Functional Regularisation for Continual Learningcited by 0
  13. 2026Amortized Acquisition Optimization for Bayesian Optimization with Variational Mutual Informationcited by 0
  14. 2026Finding Structure in Continual Learningcited by 0
  15. 2026Diffusion-Inspired Reconfiguration of Transformers for Uncertainty Calibrationcited by 0
  16. 2026Task Switching Without Forgetting via Proximal Decouplingcited by 0
  17. 2026Balancing Multimodal Learning via Functional Progresscited by 0