mlsanityData to 14 Oct 2026

Forgetting is Everywhere

arXiv 2511.04666 ↗First posted 6 Nov 2025cs.LG stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

A fundamental challenge in developing general learning algorithms is their tendency to forget past knowledge as they adapt to new data. Addressing this problem requires a principled understanding of forgetting. Yet, despite decades of study, no unified definition has emerged that offers insight into the underlying dynamics of learning. We propose an algorithm- and task-agnostic theory that characterises forgetting as a lack of self-consistency in a learner's predictive distribution, manifesting as a loss of predictive information. Our theory naturally yields a general measure of an algorithm's propensity to forget, proves that exact Bayesian inference allows for adaptation without forgetting, and provides a tautological explanation for why generative models forget when trained on their own synthetic outputs. To validate these claims, we design a comprehensive set of experiments that span classification, regression, generative modelling, and reinforcement learning. We demonstrate that forgetting is present across all deep learning settings and plays a significant role in determining learning efficiency. Together, these results establish a principled understanding of forgetting and lay the foundation for analysing and improving the information retention capabilities of general learning algorithms.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 38 of its 61 references that are in the map. First posted on arXiv 6 Nov 2025.

Builds on

38 of 61 references.

References · 38

  1. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  2. 2016Learning without Forgettingcited by 2,496
  3. 2016Progressive Neural Networkscited by 1,361
  4. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
  5. 2017Hindsight Experience Replaycited by 1,136
  6. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,105
  7. 2015Prioritized Experience Replaycited by 1,099
  8. 2018Experience Replay for Continual Learningcited by 1,095
  9. 2018Efficient Lifelong Learning with A-GEMcited by 1,058
  10. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 872
  11. 2018Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigencecited by 742
  12. 2018An Empirical Study of Example Forgetting during Deep Neural Network Learningcited by 678
26 more
  1. 2017Measuring Catastrophic Forgetting in Neural Networkscited by 360
  2. 2023The Curse of Recursion: Training on Generated Data Makes Models Forgetcited by 356
  3. 2020Towards Continual Reinforcement Learning: A Review and Perspectivescited by 274
  4. 2023Self-Consuming Generative Models Go MADcited by 250
  5. 2025RL's Razor: Why Online Reinforcement Learning Forgets Lesscited by 176
  6. 2020Revisiting Fundamentals of Experience Replaycited by 175
  7. 2018Deep Reinforcement Learning and the Deadly Triadcited by 174
  8. 2022Measuring Forgetting of Memorized Training Examplescited by 118
  9. 2018Alleviating catastrophic forgetting using context-dependent gating and synaptic stabilizationcited by 104
  10. 2023A Definition of Continual Reinforcement Learningcited by 101
  11. 2023On the Stability of Iterative Retraining of Generative Models on their own Datacited by 100
  12. 2021Continual Learning in the Teacher-Student Setup: Impact of Task Similaritycited by 88
  13. 2020Optimal Continual Learning has Perfect Memory and is NP-hardcited by 87
  14. 2018Pseudo-Rehearsal: Achieving Deep Reinforcement Learning without Catastrophic Forgettingcited by 46
  15. 2021Martingale posterior distributionscited by 45
  16. 2021Simple Agent, Complex Environment: Efficient Reinforcement Learning with Agent Statescited by 43
  17. 2023Maintaining Plasticity in Deep Continual Learningcited by 43
  18. 2023Continual Learning as Computationally Constrained Reinforcement Learningcited by 37
  19. 2022The Phenomenon of Policy Churncited by 28
  20. 2021Formalizing the Generalization-Forgetting Trade-off in Continual Learningcited by 24
  21. 2023Does Continual Learning Equally Forget All Parameters?cited by 18
  22. 2022A Theory for Knowledge Transfer in Continual Learningcited by 16
  23. 2023On the Convergence of Bounded Agentscited by 7
  24. 2025Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMscited by 7
  25. 2020Continual Learning from the Perspective of Compressioncited by 4
  26. 2021On the space of coefficients of a Feed Forward Neural Networkcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Sparsity, Superposition, and Forgetting: A Mechanistic Study of Representation Retention in Continual Learning
  2. 2026Lost or Hidden? A Concept-Level Forgetting in Supervised Continual Learning
  3. 2026To Retain or to Adapt? Generalizing Continual Learning
  4. 2026Forgetting in Language Models: Capacity, Optimization, and Self-Generated Replay
  5. 2025Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting
  6. 2022Maslow's Hammer for Catastrophic Forgetting: Node Re-Use vs Node Activation
  7. 2026Evolutionary Strategies lead to Catastrophic Forgetting in LLMs
  8. 2025Forgetting as a Feature: Cognitive Alignment of Large Language Models