An Empirical Overview of the No Free Lunch Theorem and Its Effect on Real-World Machine Learning Classification

被引：76

作者：

Gomez, David ^{[1
]}

Rojas, Alfonso ^{[1
]}

机构：

[1] Univ Politecn Cataluna, Telemat Engn Dept, Barcelona 08034, Spain

来源：

NEURAL COMPUTATION | 2016年 / 28卷 / 01期

关键词：

A-PRIORI DISTINCTIONS;

D O I：

10.1162/NECO_a_00793

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A sizable amount of research has been done to improve the mechanisms for knowledge extraction such as machine learning classification or regression. Quite unintuitively, the no free lunch (NFL) theorem states that all optimization problem strategies perform equally well when averaged over all possible problems. This fact seems to clash with the effort put forth toward better algorithms. This letter explores empirically the effect of the NFL theorem on some popular machine learning classification techniques over real-world data sets.

引用

页码：216 / 228

页数：13

共 50 条

[21] Classification versus inference learning contrasted with real-world categories
Jones, Erin L.
Ross, Brian H.
MEMORY & COGNITION, 2011, 39 (05) : 764 - 777
[22] Meaningful Machine Learning Robustness Evaluation in Real-World Machine Learning Enabled System Contexts
Hiett, Ben
Boyd, Peter
Fletcher, Charles
Gowland, Sam
Sharp, James H.
Sloggett, David
Banks, Alec
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING IN DEFENSE APPLICATIONS IV, 2022, 12276
[23] Empirical Evaluation of Federated Learning with Local Privacy for Real-World Application
Li, Paul Luo
Chai, Xiaoyu
Wadsworth, W. Duncan
Liao, Jilong
Paddock, Brandon
2020 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2020, : 1574 - 1583
[24] Learning Free Software Development from Real-World Experience
Marin Martinez, Juan Jose
2009 INTERNATIONAL CONFERENCE ON INTELLIGENT NETWORKING AND COLLABORATIVE SYSTEMS (INCOS 2009), 2009, : 417 - 420
[25] Beyond Real-world Benchmark Datasets: An Empirical Study of Node Classification with GNNs
Maekawa, Seiji
Noda, Koki
Sasaki, Yuya
Onizuka, Makoto
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
[26] Learning the precedence effect: Initial real-world tests
Wilson, K
2005 WORKSHOP ON APPLICATIONS OF SIGNAL PROCESSING TO AUDIO AND ACOUSTICS (WASPAA), 2005, : 142 - 145
[27] A Machine Learning Approach to Real-World Time to Treatment Discontinuation Prediction
Meng, Weilin
Zhang, Xinyuan
Ru, Boshu
Guan, Yuanfang
ADVANCED INTELLIGENT SYSTEMS, 2023, 5 (04)
[28] Brazilian disaster datasets and real-world instances for optimization and machine learning
Veloso, Rafaela
Cespedes, Juliana
Caunhye, Aakil
Alem, Douglas
DATA IN BRIEF, 2022, 42
[29] Machine learning fairness notions: Bridging the gap with real-world applications
Makhlouf, Karima
Zhioua, Sami
Palamidessi, Catuscia
INFORMATION PROCESSING & MANAGEMENT, 2021, 58 (05)
[30] Machine-learning algorithm in acute stroke: real-world experience
Chan, N.
Sibtain, N.
Booth, T.
de Souza, P.
Bibby, S.
Mah, Y. -h.
Teo, J.
U-King-Im, J. M.
CLINICAL RADIOLOGY, 2023, 78 (02) : E45 - E51

← 1 2 3 4 5 →