mlsanityData to 7 Oct 2026

Memory Replay GANs: learning to generate images from new categories without forgetting

arXiv 1809.02058 ↗First posted 6 Sep 2018cs.CVcited by 76

Not on the map. Why?

Abstract (arXiv)

Previous works on sequential learning address the problem of forgetting in discriminative models. In this paper we consider the case of generative models. In particular, we investigate generative adversarial networks (GANs) in the task of learning new categories in a sequential fashion. We first show that sequential fine tuning renders the network unable to properly generate images from previous categories (i.e. forgetting). Addressing this problem, we propose Memory Replay GANs (MeRGANs), a conditional GAN framework that integrates a memory replay generator. We study two methods to prevent forgetting by leveraging these replays, namely joint training with replay and replay alignment. Qualitative and quantitative experimental results in MNIST, SVHN and LSUN datasets show that our memory replay approach can generate competitive images while significantly mitigating the forgetting of previous categories.

Key facts. Cited by 76 papers in the map (data through 7 Oct 2026). First posted on arXiv 6 Sep 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 33; 2022: 15; 2023: 14; 2024: 7; 2025: 3; 2026: 4.

76 papers.

Sort:

By area:

  1. 2021Rainbow Memory: Continual Learning with a Memory of Diverse Samplescited by 273
  2. 2021Online Continual Learning in Image Classification: An Empirical Surveycited by 242
  3. 2021On Learning the Geodesic Path for Incremental Learningcited by 85
  4. 2024Continual Learning and Catastrophic Forgettingcited by 73
  5. 2021Replay in Deep Learning: Current Approaches and Missing Biological Elementscited by 71
  6. 2021ORDisCo: Effective and Efficient Usage of Incremental Unlabeled Data for Semi-supervised Continual Learningcited by 68
  7. 2021Online Continual Learning on Class Incremental Blurry Task Configuration with Anytime Inferencecited by 66
  8. 2023A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learningcited by 66
  9. 2021Continual Learning via Local Module Compositioncited by 61
  10. 2022R-DFCIL: Relation-Guided Representation Learning for Data-Free Class Incremental Learningcited by 59
  11. 2021Lifelong Person Re-Identification via Adaptive Knowledge Accumulationcited by 57
  12. 2022ACIL: Analytic Class-Incremental Learning with Absolute Memorization and Privacy Protectioncited by 56
38 more
  1. 2021Recent Advances of Continual Learning in Computer Vision: An Overviewcited by 55
  2. 2024Self-Expansion of Pre-trained Models with Mixture of Adapters for Continual Learningcited by 53
  3. 2021Continually Learning Self-Supervised Representations with Projected Functional Regularizationcited by 38
  4. 2022Uncertainty-aware Contrastive Distillation for Incremental Semantic Segmentationcited by 37
  5. 2023Learnability and Algorithm for Continual Learningcited by 28
  6. 2021GP-Tree: A Gaussian Process Classifier for Few-Shot Incremental Learningcited by 26
  7. 2021Multi-Domain Incremental Learning for Semantic Segmentationcited by 25
  8. 2022Continual Learning Based on OOD Detection and Task Maskingcited by 21
  9. 2021On the importance of cross-task features for class-incremental learningcited by 18
  10. 2021Lifelong Teacher-Student Network Learningcited by 18
  11. 2023Exploring Continual Learning of Diffusion Modelscited by 15
  12. 2021Piggyback GAN: Efficient Lifelong Learning for Image Conditioned Generationcited by 14
  13. 2021Do Not Forget to Attend to Uncertainty while Mitigating Catastrophic Forgettingcited by 12
  14. 2023Continual Evidential Deep Learning for Out-of-Distribution Detectioncited by 11
  15. 2021Continual learning in cross-modal retrievalcited by 10
  16. 2023Cost-effective On-device Continual Learning over Memory Hierarchy with Mirocited by 10
  17. 2022A Memory Transformer Network for Incremental Learningcited by 9
  18. 2022Improving Replay-Based Continual Semantic Segmentation with Smart Data Selectioncited by 8
  19. 2023Generative appearance replay for continual unsupervised domain adaptationcited by 8
  20. 2021Learning without Forgetting for 3D Point Cloud Objectscited by 7
  21. 2021Lifelong Infinite Mixture Model Based on Knowledge-Driven Dirichlet Processcited by 7
  22. 2022Online Unsupervised Domain Adaptation for Person Re-identificationcited by 7
  23. 2021Lifelong Twin Generative Adversarial Networkscited by 6
  24. 2021Continual Learning with Differential Privacycited by 6
  25. 2022Continual Learning of Dynamical Systems with Competitive Federated Reservoir Computingcited by 6
  26. 2023Restricted Orthogonal Gradient Projection for Continual Learningcited by 6
  27. 2023Model Inversion Attack via Dynamic Memory Learningcited by 6
  28. 2021Dynamic VAEs with Generative Replay for Continual Zero-shot Learningcited by 5
  29. 2021Self-Improving Semantic Perception for Indoor Localisationcited by 5
  30. 2024CLoG: Benchmarking Continual Learning of Image Generation Modelscited by 5
  31. 2022Generative Negative Replay for Continual Learningcited by 4
  32. 2024CFTS-GAN: Continual Few-Shot Teacher Student for Generative Adversarial Networkscited by 4
  33. 2021Generative Feature-driven Image Replay for Continual Learningcited by 3
  34. 2021Class Incremental Online Streaming Learningcited by 3
  35. 2023Streaming LifeLong Learning With Any-Time Inferencecited by 3
  36. 2023Remind of the Past: Incremental Learning with Analogical Promptscited by 3
  37. 2024Source-Guided Similarity Preservation for Online Person Re-Identificationcited by 3
  38. 2021The Surprising Positive Knowledge Transfer in Continual 3D Object Shape Reconstructioncited by 2

50 most cited of 76 All 76 (JSON)