迁移学习

维基百科,自由的百科全书
跳到导航 跳到搜索

迁移学习 是属于机器学习的一种研究领域。它专注于存储已有问题的解决模型,并将其利用在其他不同但相关问题上。[1] 比如说,用来辨识汽车的知识(或者是模型)也可以被用来提升识别卡车的能力。计算机领域的迁移学习和心理学常常提到的学习迁移在概念上有一定关系,但是两个领域在学术上的关系非常有限。

历史[编辑]

最早被引用的关于迁移学习的工作被认为属于Lorien Pratt。他在1993年制定了基于可辨识性的转移(DBT)算法。[2]

1997年,机器学习期刊发表了一期专门讨论迁移学习的期刊,[3] 而到了1998年,该领域已经发展到包括多任务学习,[4] 以及对其理论基础的更深入完善的分析。[5] 1998年,由Pratt和Sebastian Thrun编辑的 Learning to Learn[6]便是对该主题的回顾。

迁移学习也被应用于认知科学,比如《Connection》杂志就于1996年出版了一版特殊期刊,描述了如何通过使用迁移学习重新利用已有神经网络。[7]

应用[编辑]

迁移学习的算法基础可以源自 马尔可夫逻辑网络[8]贝叶斯网络.[9] 迁移网络还被利用与发现癌症种类 [10], 建筑物人员限额,[11] 普适智能游戏玩家,[12] 语句分类[13][14] 和 以及筛选垃圾邮件(短信)。[15]

相关概念[编辑]

  • Domain adaptation
  • General game playing
  • Multi-task learning
  • Multitask optimization


来源[编辑]

引用[编辑]

  1. ^ West, Jeremy; Ventura, Dan; Warnick, Sean. Spring Research Presentation: A Theoretical Foundation for Inductive Transfer. Brigham Young University, College of Physical and Mathematical Sciences. 2007 [2007-08-05]. (原始内容存档于2007-08-01). 
  2. ^ Pratt, L. Y. Discriminability-based transfer between neural networks (PDF). NIPS Conference: Advances in Neural Information Processing Systems 5. Morgan Kaufmann Publishers. 1993: 204–211. 
  3. ^ Pratt, L. Y.; Thrun, Sebastian. Machine Learning - Special Issue on Inductive Transfer. link.springer.com. Springer. July 1997 [2017-08-10] (英语). 
  4. ^ Caruana, R., "Multitask Learning", pp. 95-134 in Pratt & Thrun 1998
  5. ^ Baxter, J., "Theoretical Models of Learning to Learn", pp. 71-95 Pratt & Thrun 1998
  6. ^ Thrun & Pratt 2012.
  7. ^ Pratt, L. Special Issue: Reuse of Neural Networks through Transfer. Connection Science. 1996 [2017-08-10] (英语). 
  8. ^ Mihalkova, Lilyana; Huynh, Tuyen; Mooney, Raymond J., Mapping and Revising Markov Logic Networks for Transfer (PDF), Learning Proceedings of the 22nd AAAI Conference on Artificial Intelligence (AAAI-2007), Vancouver, BC: 608–614, July 2007 [2007-08-05] 
  9. ^ Niculescu-Mizil, Alexandru; Caruana, Rich, Inductive Transfer for Bayesian Network Structure Learning (PDF), Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics (AISTATS 2007), March 21–24, 2007 [2007-08-05], (原始内容存档 (PDF)于2010-06-20) 
  10. ^ Hajiramezanali, E. & Dadaneh, S. Z. & Karbalayghareh, A. & Zhou, Z. & Qian, X. Bayesian multi-domain learning for cancer subtype discovery from next-generation sequencing count data. 32nd Conference on Neural Information Processing Systems (NIPS 2018), Montréal, Canada. https://arxiv.org/pdf/1810.09433.pdf
  11. ^ Arief-Ang, I.B.; Salim, F.D.; Hamilton, M. DA-HOC: semi-supervised domain adaptation for room occupancy prediction using CO2 sensor data. 4th ACM International Conference on Systems for Energy-Efficient Built Environments (BuildSys). Delft, Netherlands: 1–10. 2017-11-08. ISBN 978-1-4503-5544-5. doi:10.1145/3137133.3137146. 
  12. ^ Banerjee, Bikramjit, and Peter Stone. "General Game Learning Using Knowledge Transfer." IJCAI. 2007.
  13. ^ Do, Chuong B.; Ng, Andrew Y. Transfer learning for text classification. Neural Information Processing Systems Foundation, NIPS*2005 (PDF). 2005 [2007-08-05]. 
  14. ^ Rajat, Raina; Ng, Andrew Y.; Koller, Daphne. Constructing Informative Priors using Transfer Learning. Twenty-third International Conference on Machine Learning (PDF). 2006 [2007-08-05]. (原始内容存档 (PDF)于2007-07-08). 
  15. ^ Bickel, Steffen. ECML-PKDD Discovery Challenge 2006 Overview. ECML-PKDD Discovery Challenge Workshop (PDF). 2006 [2007-08-05].