Learning Predictions for Algorithms with Predictions

被引:0
|
作者
Khodak, Mikhail [1 ]
Balcan, Maria-Florina [1 ]
Talwalkar, Ameet [1 ]
Vassilvitskii, Sergei [2 ]
机构
[1] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA
[2] Google Res, New York, NY USA
基金
美国国家科学基金会;
关键词
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
A burgeoning paradigm in algorithm design is the field of algorithms with predictions, in which algorithms can take advantage of a possibly-imperfect prediction of some aspect of the problem. While much work has focused on using predictions to improve competitive ratios, running times, or other performance measures, less effort has been devoted to the question of how to obtain the predictions themselves, especially in the critical online setting. We introduce a general design approach for algorithms that learn predictors: (1) identify a functional dependence of the performance measure on the prediction quality and (2) apply techniques from online learning to learn predictors, tune robustness-consistency trade-offs, and bound the sample complexity. We demonstrate the effectiveness of our approach by applying it to bipartite matching, ski-rental, page migration, and job scheduling. In several settings we improve upon multiple existing results while utilizing a much simpler analysis, while in the others we provide the first learning-theoretic guarantees.
引用
收藏
页数:14
相关论文
共 50 条
  • [21] Online Metric Algorithms with Untrusted Predictions
    Antoniadis, Antonios
    Coester, Christian
    Elias, Marek
    Polak, Adam
    Simon, Bertrand
    ACM TRANSACTIONS ON ALGORITHMS, 2023, 19 (02)
  • [22] Running time predictions for factoring algorithms
    Croot, Ernie
    Granville, Andrew
    Pemantle, Robin
    Tetali, Prasad
    ALGORITHMIC NUMBER THEORY, 2008, 5011 : 1 - +
  • [23] Augmenting Online Algorithms with ε-Accurate Predictions
    Gupta, Anupam
    Panigrahi, Debmalya
    Subercaseaux, Bernardo
    Sun, Kevin
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35, NEURIPS 2022, 2022,
  • [24] Improving the performance of machine learning algorithms for health outcomes predictions in multicentric cohorts
    Wichmann, Roberta Moreira
    Fernandes, Fernando Timoteo
    Chiavegatto Filho, Alexandre Dias Porto
    IACOV BR Network
    SCIENTIFIC REPORTS, 2023, 13 (01)
  • [25] Machine Learning Algorithms for Food Intelligence: Towards a Method for More Accurate Predictions
    Polychronou, Ioanna
    Katsivelis, Panagis
    Papakonstantinou, Mihalis
    Stoitsis, Giannis
    Manouselis, Nikos
    ENVIRONMENTAL SOFTWARE SYSTEMS: DATA SCIENCE IN ACTION, ISESS 2020, 2020, 554 : 165 - 172
  • [26] Improving the performance of machine learning algorithms for health outcomes predictions in multicentric cohorts
    Roberta Moreira Wichmann
    Fernando Timoteo Fernandes
    Alexandre Dias Porto Chiavegatto Filho
    Scientific Reports, 13
  • [27] Predictions on Predictions
    Voas, Jeffrey
    COMPUTER, 2019, 52 (12) : 80 - 82
  • [28] Comparison of machine learning and deep learning algorithms for hourly global/diffuse solar radiation predictions
    Bamisile, Olusola
    Oluwasanmi, Ariyo
    Ejiyi, Chukwuebuka
    Yimen, Nasser
    Obiora, Sandra
    Huang, Qi
    INTERNATIONAL JOURNAL OF ENERGY RESEARCH, 2022, 46 (08) : 10052 - 10073
  • [29] Deep Learning Predictions for Cryptocurrencies
    Thavaneswaran, Aerambamoorthy
    Liang, You
    Bowala, Sulalitha
    Paseka, Alex
    Ghahramani, Melody
    2022 IEEE 46TH ANNUAL COMPUTERS, SOFTWARE, AND APPLICATIONS CONFERENCE (COMPSAC 2022), 2022, : 1280 - 1285
  • [30] Implicit learning of social predictions
    Heerey, Erin A.
    Velani, Hemma
    JOURNAL OF EXPERIMENTAL SOCIAL PSYCHOLOGY, 2010, 46 (03) : 577 - 581