Safe semi-supervised learning: a brief introduction

被引:0
|
作者
Yu-Feng Li
De-Ming Liang
机构
[1] Nanjing University,National Key Laboratory for Novel Software Technology
[2] Collaborative Innovation Center of Novel Software Technology and Industrialization,undefined
来源
关键词
machine learning; semi-supervised learning; safe;
D O I
暂无
中图分类号
学科分类号
摘要
Semi-supervised learning constructs the predictive model by learning from a few labeled training examples and a large pool of unlabeled ones. It has a wide range of application scenarios and has attracted much attention in the past decades. However, it is noteworthy that although the learning performance is expected to be improved by exploiting unlabeled data, some empirical studies show that there are situations where the use of unlabeled data may degenerate the performance. Thus, it is advisable to be able to exploit unlabeled data safely. This article reviews some research progress of safe semi-supervised learning, focusing on three types of safeness issue: data quality, where the training data is risky or of low-quality; model uncertainty, where the learning algorithm fails to handle the uncertainty during training; measure diversity, where the safe performance could be adapted to diverse measures.
引用
收藏
页码:669 / 676
页数:7
相关论文
共 50 条
  • [1] Safe semi-supervised learning: a brief introduction
    Li, Yu-Feng
    Liang, De-Ming
    [J]. FRONTIERS OF COMPUTER SCIENCE, 2019, 13 (04) : 669 - 676
  • [2] Introduction to semi-supervised learning
    Goldberg, Xiaojin
    [J]. Synthesis Lectures on Artificial Intelligence and Machine Learning, 2009, 6 : 1 - 116
  • [3] Learning Safe Prediction for Semi-Supervised Regression
    Li, Yu-Feng
    Zha, Han-Wen
    Zhou, Zhi-Hua
    [J]. THIRTY-FIRST AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2017, : 2217 - 2223
  • [4] A hybrid safe semi-supervised learning method
    Gan, Haitao
    Guo, Li
    Xia, Siyu
    Wang, Tao
    [J]. EXPERT SYSTEMS WITH APPLICATIONS, 2020, 149
  • [5] Safe semi-supervised learning for pattern classification
    Ma, Jun
    Yu, Guolin
    Xiong, Weizhi
    Zhu, Xiaolong
    [J]. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2023, 121
  • [6] Dual Learning-Based Safe Semi-Supervised Learning
    Gan, Haitao
    Li, Zhenhua
    Fan, Yingle
    Luo, Zhizeng
    [J]. IEEE ACCESS, 2018, 6 : 2615 - 2621
  • [7] Safe semi-supervised learning based on weighted likelihood
    Kawakita, Masanori
    Takeuchi, Jun'ichi
    [J]. NEURAL NETWORKS, 2014, 53 : 146 - 164
  • [8] Adaptive Safe Semi-Supervised Extreme Machine Learning
    Ma, Jun
    Yuan, Chao
    [J]. IEEE ACCESS, 2019, 7 : 76176 - 76184
  • [9] Safe Semi-Supervised Learning of Sum-Product Networks
    Trapp, Martin
    Madl, Tamas
    Peharz, Robert
    Pernkopf, Franz
    Trappl, Robert
    [J]. CONFERENCE ON UNCERTAINTY IN ARTIFICIAL INTELLIGENCE (UAI2017), 2017,
  • [10] Safe semi-supervised learning using a bayesian neural network
    Bae, Jinsoo
    Lee, Minjung
    Kim, Seoung Bum
    [J]. INFORMATION SCIENCES, 2022, 612 : 453 - 464