Optimal Learners for Realizable Regression: PAC Learning and Online Learning

被引:0
|
作者
Attias, Idan [1 ]
Hanneke, Steve [2 ]
Kalavasis, Alkis [3 ]
Karbasi, Amin [3 ,4 ]
Velegkas, Grigoris [3 ]
机构
[1] Ben Gurion Univ Negev, Beer Sheva, Israel
[2] Purdue Univ, W Lafayette, IN USA
[3] Yale Univ, New Haven, CT USA
[4] Google Res, Mountain View, CA USA
关键词
UNIFORM-CONVERGENCE; LEARNABILITY; COMPRESSION; BOUNDS; PREDICTION;
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
In this work, we aim to characterize the statistical complexity of realizable regression both in the PAC learning setting and the online learning setting. Previous work had established the sufficiency of finiteness of the fat shattering dimension for PAC learnability and the necessity of finiteness of the scaled Natarajan dimension, but little progress had been made towards a more complete characterization since the work of Simon (SICOMP '97). To this end, we first introduce a minimax instance optimal learner for realizable regression and propose a novel dimension that both qualitatively and quantitatively characterizes which classes of real-valued predictors are learnable. We then identify a combinatorial dimension related to the Graph dimension that characterizes ERM learnability in the realizable setting. Finally, we establish a necessary condition for learnability based on a combinatorial dimension related to the DS dimension, and conjecture that it may also be sufficient in this context. Additionally, in the context of online learning we provide a dimension that characterizes the minimax instance optimal cumulative loss up to a constant factor and design an optimal online learner for realizable regression, thus resolving an open question raised by Daskalakis and Golowich in STOC '22.
引用
收藏
页数:33
相关论文
共 50 条
  • [1] Online PAC-Bayes Learning
    Haddouche, Maxime
    Guedj, Benjamin
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [2] Learning from the online learners
    Mishra, S
    BRITISH JOURNAL OF EDUCATIONAL TECHNOLOGY, 2005, 36 (03) : 569 - 574
  • [3] Online learners and their learning strategies
    Dewar, T
    Whittington, D
    JOURNAL OF EDUCATIONAL COMPUTING RESEARCH, 2000, 23 (04) : 385 - 403
  • [4] The Optimal Sample Complexity of PAC Learning
    Hanneke, Steve
    JOURNAL OF MACHINE LEARNING RESEARCH, 2016, 17
  • [5] Private PAC Learning May be Harder than Online Learning
    Bun, Mark
    Cohen, Aloni
    Desai, Rathin
    INTERNATIONAL CONFERENCE ON ALGORITHMIC LEARNING THEORY, VOL 237, 2024, 237
  • [6] Determinants of Learners' Self-Directed Learning and Online Learning Attitudes in Online Learning
    Li, Jing
    Wu, Chi-Hui
    SUSTAINABILITY, 2023, 15 (12)
  • [7] Engaging learners in online learning environments
    Cher Ping Lim
    TechTrends, 2004, 48 (4) : 16 - 23
  • [8] Engaging learners in synchronous online learning
    Qiyun Wang
    Yun Wen
    Choon Lang Quek
    Education and Information Technologies, 2023, 28 : 4429 - 4452
  • [9] Engaging learners in synchronous online learning
    Wang, Qiyun
    Wen, Yun
    Quek, Choon Lang
    EDUCATION AND INFORMATION TECHNOLOGIES, 2023, 28 (04) : 4429 - 4452
  • [10] Adult learners' emotions in online learning
    Zembylas, Michalinos
    DISTANCE EDUCATION, 2008, 29 (01) : 71 - 87