mlsanityData to 7 Oct 2026

Continuous Learning in Single-Incremental-Task Scenarios

arXiv 1806.08568 ↗First posted 22 Jun 2018cs.LG cs.AI cs.CVcited by 109

Not on the map. Why?

Abstract (arXiv)

It was recently shown that architectural, regularization and rehearsal strategies can be used to train deep models sequentially on a number of disjoint tasks without forgetting previously acquired knowledge. However, these strategies are still unsatisfactory if the tasks are not disjoint but constitute a single incremental task (e.g., class-incremental learning). In this paper we point out the differences between multi-task and single-incremental-task scenarios and show that well-known approaches such as LWF, EWC and SI are not ideal for incremental task scenarios. A new approach, denoted as AR1, combining architectural and regularization strategies is then specifically proposed. AR1 overhead (in term of memory and computation) is very small thus making it suitable for online learning. When tested on CORe50 and iCIFAR-100, AR1 outperformed existing regularization strategies by a good margin.

Key facts. Cited by 109 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Jun 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 36; 2022: 27; 2023: 20; 2024: 18; 2025: 7; 2026: 1.

109 papers.

Sort:

By area:

  1. 2021iMAP: Implicit Mapping and Positioning in Real-Timecited by 476
  2. 2022CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learningcited by 446
  3. 2021Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scanscited by 336
  4. 2021Online Continual Learning in Image Classification: An Empirical Surveycited by 242
  5. 2023Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampeningcited by 177
  6. 2021Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learningcited by 130
  7. 2021Avalanche: an End-to-End Library for Continual Learningcited by 124
  8. 2022The CLEAR Benchmark: Continual LEArning on Real-World Imagerycited by 92
  9. 2024MoRAL: MoE Augmented LoRA for LLMs' Lifelong Learningcited by 63
  10. 2021Continual Learning for Real-World Autonomous Systems: Algorithms, Challenges and Frameworkscited by 56
  11. 2022A Closer Look at Rehearsal-Free Continual Learningcited by 54
  12. 2021Class-Incremental Learning with Generative Classifierscited by 39
38 more
  1. 2021Online Continual Learning For Visual Food Classificationcited by 39
  2. 2024Gradient Reweighting: Towards Imbalanced Class-Incremental Learningcited by 37
  3. 2022CLAD: A realistic Continual Learning benchmark for Autonomous Drivingcited by 35
  4. 2021Continual Learning for Recurrent Neural Networks: an Empirical Evaluationcited by 34
  5. 2024Towards Efficient Replay in Federated Incremental Learningcited by 33
  6. 2021Memory-Efficient Semi-Supervised Continual Learning: The World is its Own Replay Buffercited by 30
  7. 2022Graph Lifelong Learning: A Surveycited by 30
  8. 2023Few-shot Class-incremental Learning for Classification and Object Detection: A Surveycited by 25
  9. 2021Towards Federated Learning on Time-Evolving Heterogeneous Datacited by 23
  10. 2021Is Class-Incremental Enough for Continual Learning?cited by 22
  11. 2023Task-Aware Information Routing from Common Representation Space in Lifelong Learningcited by 22
  12. 2021CLeaR: An Adaptive Continual Learning Framework for Regression Taskscited by 21
  13. 2021Elastic Weight Consolidation (EWC): Nuts and Boltscited by 21
  14. 2024Adaptive Memory Replay for Continual Learningcited by 21
  15. 2024Personalized Federated Domain-Incremental Learning based on Adaptive Knowledge Matchingcited by 21
  16. 2022Do Pre-trained Models Benefit Equally in Continual Learning?cited by 20
  17. 2023Long-Tailed Continual Learning For Visual Food Recognitioncited by 20
  18. 2021Regularization-based Continual Learning for Anomaly Detection in Discrete Manufacturingcited by 19
  19. 2021Unsupervised Continual Learning in Streaming Environmentscited by 19
  20. 2021Unsupervised Continual Learning Via Pseudo Labelscited by 18
  21. 2024Graph Learning under Distribution Shifts: A Comprehensive Survey on Domain Adaptation, Out-of-distribution, and Continual Learningcited by 17
  22. 2021Online Continual Learning Via Candidates Votingcited by 15
  23. 2024Rehearsal-Free Continual Federated Learning with Synergistic Synaptic Intelligencecited by 15
  24. 2023Prediction Error-based Classification for Class-Incremental Learningcited by 14
  25. 2021Catastrophic Forgetting in Deep Graph Networks: an Introductory Benchmark for Graph Classificationcited by 13
  26. 2022Continual Learning with Evolving Class Ontologiescited by 13
  27. 2023Overcoming the Stability Gap in Continual Learningcited by 13
  28. 2025Mitigating Visual Knowledge Forgetting in MLLM Instruction-tuning via Modality-decoupled Gradient Descentcited by 13
  29. 2023Rehearsal-Free Domain Continual Face Anti-Spoofing: Generalize More and Forget Lesscited by 12
  30. 2021Ex-Model: Continual Learning from a Stream of Trained Modelscited by 11
  31. 2022Beyond Supervised Continual Learning: a Reviewcited by 11
  32. 2023Shoggoth: Towards Efficient Edge-Cloud Collaborative Real-Time Video Inference via Adaptive Online Learningcited by 11
  33. 2024Continual Learning in the Presence of Repetitioncited by 10
  34. 2021Total Recall: a Customized Continual Learning Method for Neural Semantic Parserscited by 9
  35. 2022Memory Population in Continual Learning via Outlier Eliminationcited by 9
  36. 2022Reactive Exploration to Cope with Non-Stationarity in Lifelong Reinforcement Learningcited by 9
  37. 2023Class-Incremental Learning with Repetitioncited by 9
  38. 2024Class-incremental Learning for Time Series: Benchmark and Evaluationcited by 9

50 most cited of 109 All 109 (JSON)