谢邀。比较成熟的 子领域就不说了,这里主要介绍NLP领域内几个尚需继续更好地解决的子领域和一些较新较好的综述文.
这个其实不管是CV还是NLP领域其实都在研究,也都待进一步解决。强烈推荐 2019年来自新加坡南洋理工大学的综述长文:
Wei Wang, Vincent W. Zheng, Han Yu, and Chunyan Miao.(2019). A Survey of Zero-Shot Learning: Settings, Methods, and Applications. ACM Trans. Intell. Syst. Technol.10, 2, Article 13 (January 2019), 37 pages.
本人自己也写过一篇零样本的综述文章,可以参考下。(一种解决范式):
推荐 来自港科大和第四范式的Few-shot learning综述长文:Generalizing from a Few Examples: A Survey on Few-Shot Learning
Qi, G. J., & Luo, J. (2019). Small data challenges in big data era: A survey of recent progress on unsupervised and semi-supervised methods.arXiv preprint arXiv:1903.11260.
推荐 迁移学习领域最具代表性的综述是A survey on transfer learning,杨强老师署名的论文,虽然比较早,发表于2009-2010年,对迁移学习进行了比较权威的定义。
Pan, S. J., & Yang, Q. (2009). A survey on transfer learning. IEEE Transactions on knowledge and data engineering, 22(10), 1345-1359.
另外还有一些比较新的综述Latest survey,在这里随手介绍几篇:
[1] 2019 一篇新survey:Transfer Adaptation Learning: A Decade Survey
Zhang, L. (2019). Transfer Adaptation Learning: A Decade Survey. arXiv preprint arXiv:1903.04687.
[2] 2018 一篇迁移度量学习的综述: Transfer Metric Learning: Algorithms, Applications and Outlooks
Luo, Y., Wen, Y., Duan, L., & Tao, D. (2018). Transfer metric learning: Algorithms, applications and outlooks. arXiv preprint arXiv:1810.03944.
另外这个领域 戴老板的论文也是非常有必要读的(非综述,个人强推)
[3] 戴文渊. (2009). 基于实例和特征的迁移学习算法研究 (Doctoral dissertation, 上海: 上海交通大学).
这个比较推荐 南京大学周志华老师 的综述论文
Zhou, Z. H. (2017). A brief introduction to weakly supervised learning. National Science Review, 5(1), 44-53.
2019 google的T5模型论文,把它当成综述来看就介绍的挺好:
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., ... & Liu, P. J. (2019). Exploring the limits of transfer learning with a unified text-to-text transformer. arXiv preprint arXiv:1910.10683.
bert后还有一些改进模型比如华为刘群/百度的ERNIE,XLNet等相关非综述文章,可以自行阅读。
还有一些比较新的不同方向的综述文:
[1] 注意力机制:Hu, D. (2019, September). An introductory survey on attention mechanisms in nlp problems. In Proceedings of SAI Intelligent Systems Conference (pp. 432-448). Springer, Cham.
[2] Elvis Saravia and Soujanya:PoriaElvis Saravia and Soujanya Poria:NLP方方面面都有涉及,颇有一些横贯全局的意思。
网址:
这里只是分不同研究方向列举了一些,其余相关论文可以参考本文文章:
https://zhuanlan.zhihu.com/p/91408237;
================================================
增补一些20200419:
1) Zhi-Hua Zhou, A brief introduction to weakly supervised learning, National Science Review, Volume 5, Issue 1, January 2018, Pages 44–53, https://doi.org/10.1093/nsr/nwx106;
关于课程学习的文章:
2) Bengio, Y., Louradour, J., Collobert, R., Weston, J.: Curriculum learning. In: ICML. pp. 41–48. ACM (2009)
3) Luo, B., Feng, Y., Wang, Z., Zhu, Z., Huang, S., Yan, R., & Zhao, D. (2017). Learning with noise: Enhance distantly supervised relation extraction with dynamic transition matrix. arXiv preprint arXiv:1705.03995.
4) Guo, S., Huang, W., Zhang, H., Zhuang, C., Dong, D., Scott, M. R., & Huang, D. (2018). Curriculumnet: Weakly supervised learning from large-scale web images. In Proceedings of the European Conference on Computer Vision (ECCV) (pp. 135-150).
[1] 陈丹琦博士的毕业论文 Chen, D. (2018).Neural Reading Comprehension and Beyond(Doctoral dissertation, Stanford University).
[2] 来自国防科技大学的综述长文 Liu, S., Zhang, X., Zhang, S., Wang, H., & Zhang, W. (2019). Neural machine reading comprehension: Methods and trends.Applied Sciences,9(18), 3698.
更多可见:
女王:求求题主放过我,我可不敢有什么政绩。。。
女王:求求题主放过我,我可不敢有什么政绩。。。
女王:求求题主放过我,我可不敢有什么政绩。。。