On Belief Change for Multi-Label Classifier Encodings

被引:0
|
作者
Coste-Marquis, Sylvie [1 ,2 ]
Marquis, Pierre [1 ,2 ,3 ]
机构
[1] Univ Artois, CRIL, Arras, France
[2] CNRS, Arras, France
[3] Inst Univ France, Paris, France
关键词
REVISION;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
An important issue in ML consists in developing approaches exploiting background knowledge T for improving the accuracy and the robustness of learned classifiers C. Delegating the classification task to a Boolean circuit Sigma exhibiting the same input-output behaviour as C, the problem of exploiting T within C can be viewed as a belief change scenario. However, usual change operations are not suited to the task of modifying the classifier encoding Sigma in a minimal way, to make it complying with T. To fill the gap, we present a new belief change operation, called rectification. We characterize the family of rectification operators from an axiomatic perspective and exhibit operators from this family. We identify the standard belief change postulates that every rectification operator satisfies and those it does not. We also focus on some computational aspects of rectification and compliance.
引用
收藏
页码:1829 / 1836
页数:8
相关论文
共 50 条
  • [1] Classifier chains for multi-label classification
    Jesse Read
    Bernhard Pfahringer
    Geoff Holmes
    Eibe Frank
    Machine Learning, 2011, 85
  • [2] Classifier Chains for Multi-label Classification
    Read, Jesse
    Pfahringer, Bernhard
    Holmes, Geoff
    Frank, Eibe
    MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, PT II, 2009, 5782 : 254 - 269
  • [3] Classifier chains for multi-label classification
    Read, Jesse
    Pfahringer, Bernhard
    Holmes, Geoff
    Frank, Eibe
    MACHINE LEARNING, 2011, 85 (03) : 333 - 359
  • [4] Evolving multi-label fuzzy classifier
    Lughofer, Edwin
    INFORMATION SCIENCES, 2022, 597 : 1 - 23
  • [5] A Multi-Label Classifier for Online Petition Systems
    Buryakov, Daniil
    Kovacs, Mate
    Serdult, Uwe
    Kryssanov, Victor
    PROCEEDINGS OF THE 25TH ANNUAL INTERNATIONAL CONFERENCE ON DIGITAL GOVERNMENT RESEARCH, DGO 2024, 2024, : 156 - 164
  • [6] Classifier Selection Approaches for Multi-label Problems
    Pillai, Ignazio
    Fumera, Giorgio
    Roli, Fabio
    MULTIPLE CLASSIFIER SYSTEMS, 2011, 6713 : 167 - 176
  • [7] Dynamic Classifier Chains for Multi-label Learning
    Trajdos, Pawel
    Kurzynski, Marek
    PATTERN RECOGNITION, DAGM GCPR 2019, 2019, 11824 : 567 - 580
  • [8] On the Optimality of Classifier Chain for Multi-label Classification
    Liu, Weiwei
    Tsang, Ivor W.
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 28 (NIPS 2015), 2015, 28
  • [9] Classifier circle method for multi-label learning
    Key Laboratory for Novel Software Technology, Nanjing University, Nanjing
    210023, China
    Ruan Jian Xue Bao, 11 (2811-2819):
  • [10] Identification of Multi-Functional Enzyme with Multi-Label Classifier
    Che, Yuxin
    Ju, Ying
    Xuan, Ping
    Long, Ren
    Xing, Fei
    PLOS ONE, 2016, 11 (04):