Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
80 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
7 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Saliency-Guided Hidden Associative Replay for Continual Learning (2310.04334v1)

Published 6 Oct 2023 in cs.LG

Abstract: Continual Learning is a burgeoning domain in next-generation AI, focusing on training neural networks over a sequence of tasks akin to human learning. While CL provides an edge over traditional supervised learning, its central challenge remains to counteract catastrophic forgetting and ensure the retention of prior tasks during subsequent learning. Amongst various strategies to tackle this, replay based methods have emerged as preeminent, echoing biological memory mechanisms. However, these methods are memory intensive, often preserving entire data samples, an approach inconsistent with humans selective memory retention of salient experiences. While some recent works have explored the storage of only significant portions of data in episodic memory, the inherent nature of partial data necessitates innovative retrieval mechanisms. Current solutions, like inpainting, approximate full data reconstruction from partial cues, a method that diverges from genuine human memory processes. Addressing these nuances, this paper presents the Saliency Guided Hidden Associative Replay for Continual Learning. This novel framework synergizes associative memory with replay-based strategies. SHARC primarily archives salient data segments via sparse memory encoding. Importantly, by harnessing associative memory paradigms, it introduces a content focused memory retrieval mechanism, promising swift and near-perfect recall, bringing CL a step closer to authentic human memory processes. Extensive experimental results demonstrate the effectiveness of our proposed method for various continual learning tasks.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (39)
  1. Gradient based sample selection for online continual learning. arXiv preprint arXiv:1903.08671, 2019.
  2. Modeling brain function: The world of attractor neural networks. Cambridge university press, 1989.
  3. Learning fast, learning slow: A general continual learning method based on complementary learning system. arXiv preprint arXiv:2201.12604, 2022.
  4. Saliency-augmented memory completion for continual learning. In Proceedings of the 2023 SIAM International Conference on Data Mining (SDM), pages 244–252. SIAM, 2023.
  5. G. Bai and L. Zhao. Saliency-regularized deep multi-task learning. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 15–25, 2022.
  6. Dark experience for general continual learning: a strong, simple baseline. Advances in neural information processing systems, 33:15920–15930, 2020.
  7. Efficient lifelong learning with a-gem. arXiv preprint arXiv:1812.00420, 2018.
  8. On tiny episodic memories in continual learning. arXiv preprint arXiv:1902.10486, 2019.
  9. Remembering for the right reasons: Explanations reduce catastrophic forgetting. Applied AI letters, 2(4):e44, 2021.
  10. Image inpainting: A review. Neural Processing Letters, 51:2007–2028, 2020.
  11. Res: A robust framework for guiding visual explanation. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 432–442, 2022.
  12. Generative adversarial networks. Communications of the ACM, 63(11):139–144, 2020.
  13. Remind your neural network to prevent catastrophic forgetting. In European Conference on Computer Vision, pages 466–483. Springer, 2020.
  14. D. Hebb. The organization of behavior. New York, 1949.
  15. J. J. Hopfield. Neural networks and physical systems with emergent collective computational abilities. Proceedings of the national academy of sciences, 79(8):2554–2558, 1982.
  16. Y. Huang and R. P. Rao. Predictive coding. Wiley Interdisciplinary Reviews: Cognitive Science, 2(5):580–593, 2011.
  17. The nonhuman primate hippocampus: neuroanatomy and patterns of cortical connectivity. The hippocampus from cells to systems: Structure, connectivity, and functional contributions to memory and flexible cognition, pages 3–36, 2017.
  18. D. Ji and M. A. Wilson. Coordinated memory replay in the visual cortex and hippocampus during sleep. Nature neuroscience, 10(1):100–107, 2007.
  19. Overcoming catastrophic forgetting in neural networks. Proceedings of the national academy of sciences, 114(13):3521–3526, 2017.
  20. D. Krotov and J. Hopfield. Large associative memory problem in neurobiology and machine learning. arXiv preprint arXiv:2008.06996, 2020.
  21. Z. Li and D. Hoiem. Learning without forgetting. IEEE transactions on pattern analysis and machine intelligence, 40(12):2935–2947, 2017.
  22. D. Lopez-Paz and M. Ranzato. Gradient episodic memory for continual learning. Advances in neural information processing systems, 30:6467–6476, 2017.
  23. M. McCloskey and N. J. Cohen. Catastrophic interference in connectionist networks: The sequential learning problem. In Psychology of learning and motivation, volume 24, pages 109–165. Elsevier, 1989.
  24. Continual lifelong learning with neural networks: A review. Neural Networks, 113:54–71, 2019.
  25. Generating diverse structure for image inpainting with hierarchical vq-vae. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10775–10784, 2021.
  26. Variational autoencoder for deep learning of images, labels and captions. Advances in neural information processing systems, 29:2352–2360, 2016.
  27. Hopfield networks is all you need. arXiv preprint arXiv:2008.02217, 2020.
  28. Predictive coding in the visual cortex: a functional interpretation of some extra-classical receptive-field effects. Nature neuroscience, 2(1):79–87, 1999.
  29. icarl: Incremental classifier and representation learning. In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, pages 2001–2010, 2017.
  30. Learning to learn without forgetting by maximizing transfer and minimizing interference. arXiv preprint arXiv:1810.11910, 2018.
  31. A. Robins. Catastrophic forgetting, rehearsal and pseudorehearsal. Connection Science, 7(2):123–146, 1995.
  32. E. T. Rolls. The mechanisms for pattern completion and pattern separation in the hippocampus. Frontiers in systems neuroscience, 7:74, 2013.
  33. G. Saha and K. Roy. Saliency guided experience packing for replay in continual learning. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 5273–5283, 2023.
  34. Associative memories via predictive coding. Advances in Neural Information Processing Systems, 34:3874–3886, 2021.
  35. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision, pages 618–626, 2017.
  36. Deeply learned face representations are sparse, selective, and robust. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 2892–2900, 2015.
  37. The hippocampal indexing theory and episodic memory: updating the index. Hippocampus, 17(12):1158–1169, 2007.
  38. J. Yoo and F. Wood. Bayespcn: A continually learnable predictive coding associative memory. Advances in Neural Information Processing Systems, 35:29903–29914, 2022.
  39. Lifelong object detection. arXiv preprint arXiv:2009.01129, 2020.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (5)
  1. Guangji Bai (24 papers)
  2. Qilong Zhao (3 papers)
  3. Xiaoyang Jiang (3 papers)
  4. Yifei Zhang (167 papers)
  5. Liang Zhao (353 papers)
Citations (5)
View on Semantic Scholar