mlsanityData to 7 Oct 2026

On Tiny Episodic Memories in Continual Learning

arXiv 1902.10486 ↗First posted 27 Feb 2019cs.LG stat.MLcited by 299

Not on the map. Why?

Abstract (arXiv)

In continual learning (CL), an agent learns from a stream of tasks leveraging prior experience to transfer knowledge to future tasks. It is an ideal framework to decrease the amount of supervision in the existing learning algorithms. But for a successful knowledge transfer, the learner needs to remember how to perform previous tasks. One way to endow the learner the ability to perform tasks seen in the past is to store a small memory, dubbed episodic memory, that stores few examples from previous tasks and then to replay these examples when training for future tasks. In this work, we empirically analyze the effectiveness of a very small episodic memory in a CL setup where each training example is only seen once. Surprisingly, across four rather different supervised learning benchmarks adapted to CL, a very simple baseline, that jointly trains on both examples from the current task as well as examples stored in the episodic memory, significantly outperforms specifically designed CL approaches with and without episodic memory. Interestingly, we find that repetitive training on even tiny memories of past tasks does not harm generalization, on the contrary, it improves it, with gains between 7% and 17% when the memory is populated with a single example per class.

Key facts. Cited by 299 papers in the map (data through 7 Oct 2026). First posted on arXiv 27 Feb 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 29; 2022: 60; 2023: 64; 2024: 65; 2025: 45; 2026: 36.

299 papers.

Sort:

By area:

  1. 2022CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learningcited by 446
  2. 2021Gradient Projection Memory for Continual Learningcited by 338
  3. 2023Orthogonal Subspace Learning for Language Model Continual Learningcited by 296
  4. 2022S-Prompts Learning with Pre-trained Transformers: An Occam's Razor for Domain Incremental Learningcited by 246
  5. 2021New Insights on Reducing Abrupt Representation Change in Online Continual Learningcited by 215
  6. 2022A Model or 603 Exemplars: Towards Memory-Efficient Class-Incremental Learningcited by 152
  7. 2022NOTE: Robust Continual Test-time Adaptation Against Temporal Correlationcited by 150
  8. 2023Continual Diffusion: Continual Customization of Text-to-Image Diffusion with C-LoRAcited by 138
  9. 2021Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learningcited by 130
  10. 2021Online Coreset Selection for Rehearsal-based Continual Learningcited by 114
  11. 2021Representational Continuity for Unsupervised Continual Learningcited by 95
  12. 2022The CLEAR Benchmark: Continual LEArning on Real-World Imagerycited by 92
38 more
  1. 2022TRGP: Trust Region Gradient Projection for Continual Learningcited by 79
  2. 2021Rehearsal revealed: The limits and merits of revisiting samples in continual learningcited by 76
  3. 2022Probing Representation Forgetting in Supervised and Unsupervised Continual Learningcited by 74
  4. 2021Flattening Sharpness for Dynamic Gradient Projection Memory Benefits Continual Learningcited by 71
  5. 2023A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learningcited by 66
  6. 2021Continual Learning via Local Module Compositioncited by 61
  7. 2023Computationally Budgeted Continual Learning: What Does Matter?cited by 61
  8. 2022Beyond Not-Forgetting: Continual Learning with Backward Knowledge Transfercited by 59
  9. 2022Continual evaluation for lifelong learning: Identifying the stability gapcited by 55
  10. 2022A Closer Look at Rehearsal-Free Continual Learningcited by 54
  11. 2022An Efficient Domain-Incremental Learning Approach to Drive in All Weather Conditionscited by 54
  12. 2023Real-Time Evaluation in Online Continual Learning: A New Hopecited by 47
  13. 2021Posterior Meta-Replay for Continual Learningcited by 45
  14. 2022A Continual Deepfake Detection Benchmark: Dataset, Methods, and Essentialscited by 45
  15. 2022Memory Efficient Continual Learning with Transformerscited by 43
  16. 2022Class-incremental Novel Class Discoverycited by 42
  17. 2024SDDGR: Stable Diffusion-based Deep Generative Replay for Class Incremental Object Detectioncited by 42
  18. 2022NISPA: Neuro-Inspired Stability-Plasticity Adaptation for Continual Learning in Sparse Networkscited by 38
  19. 2021Non-IID data and Continual Learning processes in Federated Learning: A long road aheadcited by 36
  20. 2025CL-MoE: Enhancing Multimodal Large Language Model with Dual Momentum Mixture-of-Experts for Continual Visual Question Answeringcited by 36
  21. 2023Continual Learning as Computationally Constrained Reinforcement Learningcited by 35
  22. 2023Prototype-Sample Relation Distillation: Towards Replay-Free Continual Learningcited by 34
  23. 2023Continual Learning with Scaled Gradient Projectioncited by 33
  24. 2024Convolutional Prompting meets Language Models for Continual Learningcited by 32
  25. 2022Disentangling Transfer in Continual Reinforcement Learningcited by 31
  26. 2021Memory-Efficient Semi-Supervised Continual Learning: The World is its Own Replay Buffercited by 30
  27. 2022Meta-attention for ViT-backed Continual Learningcited by 29
  28. 2022Improving Task-free Continual Learning by Distributionally Robust Memory Evolutioncited by 29
  29. 2022Balancing Stability and Plasticity through Advanced Null Space in Continual Learningcited by 29
  30. 2023Online Continual Learning Without the Storage Constraintcited by 29
  31. 2023CTP: Towards Vision-Language Continual Pretraining via Compatible Momentum Contrast and Topology Preservationcited by 29
  32. 2023Learnability and Algorithm for Continual Learningcited by 28
  33. 2023Diversifying the Mixture-of-Experts Representation for Language Models with Orthogonal Optimizercited by 28
  34. 2022Learning Bayesian Sparse Networks with Full Experience Replay for Continual Learningcited by 27
  35. 2022A Multi-Head Model for Continual Learning via Out-of-Distribution Replaycited by 27
  36. 2024Bi-Factorial Preference Optimization: Balancing Safety-Helpfulness in Language Modelscited by 27
  37. 2025Dynamic Mixture of Curriculum LoRA Experts for Continual Multimodal Instruction Tuningcited by 27
  38. 2024Online-LoRA: Task-free Online Continual Learning via Low Rank Adaptationcited by 26

50 most cited of 299 All 299 (JSON)