mlsanityData to 14 Oct 2026

Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interference

arXiv 1810.11910 ↗First posted 29 Oct 2018cs.LG cs.AI stat.MLcited by 521

Not on the map. Why?

Abstract (arXiv)

Lack of performance when it comes to continual learning over non-stationary distributions of data remains a major challenge in scaling neural network learning to more human realistic settings. In this work we propose a new conceptualization of the continual learning problem in terms of a temporally symmetric trade-off between transfer and interference that can be optimized by enforcing gradient alignment across examples. We then propose a new algorithm, Meta-Experience Replay (MER), that directly exploits this view by combining experience replay with optimization based meta-learning. This method learns parameters that make interference based on future gradients less likely and transfer based on future gradients more likely. We conduct experiments across continual lifelong supervised learning benchmarks and non-stationary reinforcement learning environments demonstrating that our approach consistently outperforms recently proposed baselines for continual learning. Our experiments show that the gap between the performance of MER and baseline algorithms grows both as the environment gets more non-stationary and as the fraction of the total experiences stored gets smaller.

Key facts. Cited by 521 papers in the map (data through 14 Oct 2026). First posted on arXiv 29 Oct 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 87; 2022: 106; 2023: 99; 2024: 97; 2025: 78; 2026: 54.

521 papers.

Sort:

By area:

  1. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 872
  2. 2023An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuningcited by 639
  3. 2021Gradient Projection Memory for Continual Learningcited by 348
  4. 2022Prompt-aligned Gradient for Prompt Tuningcited by 341
  5. 2021Rainbow Memory: Continual Learning with a Memory of Diverse Samplescited by 273
  6. 2022S-Prompts Learning with Pre-trained Transformers: An Occam's Razor for Domain Incremental Learningcited by 246
  7. 2021Online Continual Learning in Image Classification: An Empirical Surveycited by 245
  8. 2021Co2L: Contrastive Continual Learningcited by 244
  9. 2024Continual Learning of Large Language Models: A Comprehensive Surveycited by 234
  10. 2023Class-Incremental Learning: A Surveycited by 209
  11. 2021Few-Shot Incremental Learning with Continually Evolved Classifierscited by 187
  12. 2023Mitigating the Alignment Tax of RLHFcited by 183
38 more
  1. 2021Supervised Contrastive Replay: Revisiting the Nearest Class Mean Classifier in Online Class-Incremental Continual Learningcited by 141
  2. 2021An Empirical Investigation of the Role of Pre-training in Lifelong Learningcited by 134
  3. 2022Continual Learning and Private Unlearningcited by 130
  4. 2024Simple and Scalable Strategies to Continually Pre-train Large Language Modelscited by 127
  5. 2022Class-Incremental Continual Learning into the eXtended DER-versecited by 124
  6. 2021Overcoming Catastrophic Forgetting in Incremental Few-Shot Learning by Finding Flat Minimacited by 118
  7. 2021Semantic-aware Knowledge Distillation for Few-Shot Class-Incremental Learningcited by 115
  8. 2022Learning Fast, Learning Slow: A General Continual Learning Method based on Complementary Learning Systemcited by 112
  9. 2022BOME! Bilevel Optimization Made Easy: A Simple First-Order Approachcited by 105
  10. 2022Remember the Past: Distilling Datasets into Addressable Memories for Neural Networkscited by 104
  11. 2021GCR: Gradient Coreset Based Replay Buffer Selection For Continual Learningcited by 95
  12. 2022The CLEAR Benchmark: Continual LEArning on Real-World Imagerycited by 93
  13. 2021Achieving Forgetting Prevention and Knowledge Transfer in Continual Learningcited by 91
  14. 2021Continual Learning for Text Classification with Information Disentanglement Based Regularizationcited by 86
  15. 2021Invariant Information Bottleneck for Domain Generalizationcited by 86
  16. 2022Sparse Mixture-of-Experts are Domain Generalizable Learnerscited by 85
  17. 2022Continual Learning of Natural Language Processing Tasks: A Surveycited by 83
  18. 2022TRGP: Trust Region Gradient Projection for Continual Learningcited by 82
  19. 2024Continual Learning and Catastrophic Forgettingcited by 77
  20. 2021AFEC: Active Forgetting of Negative Transfer in Continual Learningcited by 73
  21. 2021Replay in Deep Learning: Current Approaches and Missing Biological Elementscited by 72
  22. 2021Flattening Sharpness for Dynamic Gradient Projection Memory Benefits Continual Learningcited by 72
  23. 2023Theory on Forgetting and Generalization of Continual Learningcited by 72
  24. 2023A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learningcited by 69
  25. 2023RMM: Reinforced Memory Management for Class-Incremental Learningcited by 66
  26. 2023TARGET: Federated Class-Continual Learning via Exemplar-Free Distillationcited by 65
  27. 2023Advances and Challenges in Meta-Learning: A Technical Reviewcited by 65
  28. 2021Wide Neural Networks Forget Less Catastrophicallycited by 64
  29. 2025SD-LoRA: Scalable Decoupled Low-Rank Adaptation for Class Incremental Learningcited by 63
  30. 2024CoIN: A Benchmark of Continual Instruction tuNing for Multimodel Large Language Modelcited by 61
  31. 2022Beyond Not-Forgetting: Continual Learning with Backward Knowledge Transfercited by 59
  32. 2022Continual evaluation for lifelong learning: Identifying the stability gapcited by 56
  33. 2021Recent Advances of Continual Learning in Computer Vision: An Overviewcited by 55
  34. 2021Model Zoo: A Growing "Brain" That Learns Continuallycited by 54
  35. 2023Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomycited by 51
  36. 2023A Survey on Few-Shot Class-Incremental Learningcited by 48
  37. 2022Architecture Matters in Continual Learningcited by 46
  38. 2024ConFIG: Towards Conflict-free Training of Physics Informed Neural Networkscited by 46

50 most cited of 521 All 521 (JSON)