mlsanityData to 14 Oct 2026

Selective Replay Enhances Learning in Online Continual Analogical Reasoning

arXiv 2103.03987 ↗First posted 6 Mar 2021cs.AI cs.CV cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

In continual learning, a system learns from non-stationary data streams or batches without catastrophic forgetting. While this problem has been heavily studied in supervised image classification and reinforcement learning, continual learning in neural networks designed for abstract reasoning has not yet been studied. Here, we study continual learning of analogical reasoning. Analogical reasoning tests such as Raven's Progressive Matrices (RPMs) are commonly used to measure non-verbal abstract reasoning in humans, and recently offline neural networks for the RPM problem have been proposed. In this paper, we establish experimental baselines, protocols, and forward and backward transfer metrics to evaluate continual learners on RPMs. We employ experience replay to mitigate catastrophic forgetting. Prior work using replay for image classification tasks has found that selectively choosing the samples to replay offers little, if any, benefit over random selection. In contrast, we find that selective replay can significantly outperform random selection for the RPM task.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Builds on 54 of its 88 references that are in the map. First posted on arXiv 6 Mar 2021.

Builds on

54 of 88 references.

References · 54

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  2. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  3. 2015VQA: Visual Question Answeringcited by 2,506
  4. 2016Learning without Forgettingcited by 2,496
  5. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,333
  6. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  7. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
  8. 2017Continual Learning with Deep Generative Replaycited by 1,158
  9. 2017Hindsight Experience Replaycited by 1,136
  10. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,105
  11. 2015Prioritized Experience Replaycited by 1,099
  12. 2018Efficient Lifelong Learning with A-GEMcited by 1,058
38 more
  1. 2018Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigencecited by 742
  2. 2019Large Scale Incremental Learningcited by 732
  3. 2017Lifelong Learning with Dynamically Expandable Networkscited by 694
  4. 2018Overcoming catastrophic forgetting with hard attention to the taskcited by 681
  5. 2018End-to-End Incremental Learningcited by 597
  6. 2019Gradient based sample selection for online continual learningcited by 571
  7. 2019Online Continual Learning with Maximally Interfered Retrievalcited by 379
  8. 2017Measuring Catastrophic Forgetting in Neural Networkscited by 360
  9. 2017A simple neural network module for relational reasoningcited by 346
  10. 2019Analysing Mathematical Reasoning Abilities of Neural Modelscited by 335
  11. 2018Learning without Memorizingcited by 250
  12. 2019RAVEN: A Dataset for Relational and Analogical Visual rEasoNingcited by 226
  13. 2020Mnemonics Training: Multi-Class Incremental Learning without Forgettingcited by 221
  14. 2016Attend, Infer, Repeat: Fast Scene Understanding with Generative Modelscited by 209
  15. 2017Incremental Learning of Object Detectors without Catastrophic Forgettingcited by 208
  16. 2018Online Structured Laplace Approximations For Overcoming Catastrophic Forgettingcited by 201
  17. 2019REMIND Your Neural Network to Prevent Catastrophic Forgettingcited by 197
  18. 2020Modeling the Background for Incremental Learning in Semantic Segmentationcited by 190
  19. 2019Learning to Remember: A Synaptic Plasticity Driven Framework for Continual Learningcited by 182
  20. 2017FearNet: Brain-Inspired Model for Incremental Learningcited by 177
  21. 2018Measuring abstract reasoning in neural networkscited by 166
  22. 2019Incremental Learning Techniques for Semantic Segmentationcited by 141
  23. 2018Memory Efficient Experience Replay for Streaming Learningcited by 128
  24. 2019A continual learning survey: Defying forgetting in classification taskscited by 118
  25. 2018Don't forget, there is more than forgetting: new metrics for Continual Learningcited by 115
  26. 2014A Multi-World Approach to Question Answering about Real-World Scenes based on Uncertain Inputcited by 112
  27. 2015Grounding of Textual Phrases in Images by Reconstructioncited by 112
  28. 2020A Comprehensive Study of Class Incremental Learning Algorithms for Visual Taskscited by 107
  29. 2019Lifelong Machine Learning with Deep Streaming Linear Discriminant Analysiscited by 92
  30. 2020Stratified Rule-Aware Network for Abstract Visual Reasoningcited by 72
  31. 2019Learning Perceptual Inference by Contrastingcited by 57
  32. 2016Visual Question Answering: Datasets, Algorithms, and Future Challengescited by 56
  33. 2019Learning to Make Analogies by Contrasting Abstract Relational Structurecited by 53
  34. 2020Abstract Diagrammatic Reasoning with Multiplex Graph Networkscited by 37
  35. 2016Automatic Description Generation from Images: A Survey of Models, Datasets, and Evaluation Measurescited by 36
  36. 2020RODEO: Replay for Online Object Detectioncited by 35
  37. 2019Psycholinguistics meets Continual Learning: Measuring Catastrophic Forgetting in Visual Question Answeringcited by 33
  38. 2019Abstract Reasoning with Distracting Featurescited by 33

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 3; 2023: 5; 2024: 3.

12 papers.

  1. 2021Recent Advances of Continual Learning in Computer Vision: An Overviewcited by 55
  2. 2023The Ideal Continual Learner: An Agent That Never Forgetscited by 35
  3. 2024Adaptive Memory Replay for Continual Learningcited by 22
  4. 2022Multimodal Analogical Reasoning over Knowledge Graphscited by 21
  5. 2023GRASP: A Rehearsal Policy for Efficient Online Continual Learningcited by 14
  6. 2023Overcoming the Stability Gap in Continual Learningcited by 13
  7. 2024Learning to Route for Dynamic Adapter Composition in Continual Learning with Language Modelscited by 11
  8. 2023Continual Learning for Predictive Maintenance: Overview and Challengescited by 10
  9. 2022Memory Population in Continual Learning via Outlier Eliminationcited by 9
  10. 2022How Relevant is Selective Memory Population in Lifelong Language Learning?cited by 6
  11. 2023Masked Autoencoders are Efficient Continual Federated Learnerscited by 1
  12. 2024Data-Incremental Continual Offline Reinforcement Learningcited by 0