MetaGrad: Adaptation using Multiple Learning Rates in Online Learning

被引:0
|
作者
van Erven, Tim [1 ]
Koolen, Wouter M. [2 ]
van der Hoeven, Dirk [3 ]
机构
[1] Univ Amsterdam, Korteweg Vries Inst Math, Sci Pk 107, NL-1098 XG Amsterdam, Netherlands
[2] Ctr Wiskunde & Informat, Sci Pk 123, NL-1098 XG Amsterdam, Netherlands
[3] Leiden Univ, Math Inst, Niels Bohrweg 1, NL-2300 RA Leiden, Netherlands
关键词
online convex optimization; adaptivity; FREQUENT DIRECTIONS;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We provide a new adaptive method for online convex optimization, MetaGrad, that is robust to general convex losses but achieves faster rates for a broad class of special functions, including exp-concave and strongly convex functions, but also various types of stochastic and non-stochastic functions without any curvature. We prove this by drawing a connection to the Bernstein condition, which is known to imply fast rates in offline statistical learning. MetaGrad further adapts automatically to the size of the gradients. Its main feature is that it simultaneously considers multiple learning rates, which are weighted directly proportional to their empirical performance on the data using a new meta-algorithm. We provide three versions of MetaGrad. The full matrix version maintains a full covariance matrix and is applicable to learning tasks for which we can afford update time quadratic in the dimension. The other two versions provide speed-ups for high-dimensional learning tasks with an update time that is linear in the dimension: one is based on sketching, the other on running a separate copy of the basic algorithm per coordinate. We evaluate all versions of MetaGrad on benchmark online classification and regression tasks, on which they consistently outperform both online gradient descent and AdaGrad.
引用
收藏
页数:61
相关论文
共 50 条
  • [21] Online Adaptation for Enhancing Imitation Learning Policies
    Malato, Federico
    Hautam, Ville
    2024 IEEE CONFERENCE ON GAMES, COG 2024, 2024,
  • [22] Adaptation of Learning Strategies in Learning Objects for using Learning Styles
    Rojas Moreno, Javier Enrique
    INTERNATIONAL CONFERENCE ON FUTURE COMPUTER SUPPORTED EDUCATION, 2012, 2 : 808 - 814
  • [23] Fast Rates for Nonparametric Online Learning: From Realizability to Learning in Games
    Daskalakis, Constantinos
    Golowich, Noah
    PROCEEDINGS OF THE 54TH ANNUAL ACM SIGACT SYMPOSIUM ON THEORY OF COMPUTING (STOC '22), 2022, : 846 - 859
  • [24] Strong Convex Loss Can Increase the Learning Rates of Online Learning
    Sheng, Baohuai
    Duan, Liqin
    Ye, Peixin
    JOURNAL OF COMPUTERS, 2014, 9 (07) : 1606 - 1611
  • [25] Online Learning: Custom Design to Promote Learning for Multiple Disciplines
    Silverstone, S.
    Phadungtin, J.
    PROCEEDINGS OF WORLD ACADEMY OF SCIENCE, ENGINEERING AND TECHNOLOGY, VOL 28, 2008, 28 : 248 - +
  • [26] Online Multiple Instance Learning with No Regret
    Li, Mu
    Kwok, James T.
    Lu, Bao-Liang
    2010 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2010, : 1395 - 1401
  • [27] Distributed Online Learning With Multiple Kernels
    Hong, Songnam
    Chae, Jeongmin
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2023, 34 (03) : 1263 - 1277
  • [28] Online identification and control of a DC motor using learning adaptation of neural networks
    Rubaai, A
    Kotaru, R
    IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, 2000, 36 (03) : 935 - 942
  • [29] Online Learning with (Multiple) Kernels: A Review
    Diethe, Tom
    Girolami, Mark
    NEURAL COMPUTATION, 2013, 25 (03) : 567 - 625
  • [30] Multiple object tracking using incremental learning for appearance model adaptation
    Pernkopf, Franz
    VISAPP 2008: PROCEEDINGS OF THE THIRD INTERNATIONAL CONFERENCE ON COMPUTER VISION THEORY AND APPLICATIONS, VOL 2, 2008, : 463 - 468