Large-scale probabilistic predictors with and without guarantees of validity

被引：0

作者：

Vovk, Vladimir ^{[1
]}

Petej, Ivan ^{[1
]}

Fedorova, Valentina ^{[2
]}

机构：

[1] Univ London, Dept Comp Sci, Royal Holloway, Egham, Surrey, England

[2] Yandex, Moscow, Russia

来源：

ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 28 (NIPS 2015) | 2015年 / 28卷

基金：

英国工程与自然科学研究理事会;

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：

This paper studies theoretically and empirically a method of turning machine-learning algorithms into probabilistic predictors that automatically enjoys a property of validity (perfect calibration) and is computationally efficient. The price to pay for perfect calibration is that these probabilistic predictors produce imprecise (in practice, almost precise for large data sets) probabilities. When these imprecise probabilities are merged into precise probabilities, the resulting predictors, while losing the theoretical property of perfect calibration, are consistently more accurate than the existing methods in empirical studies.

引用

页数：9

共 50 条

[21] PET: Probabilistic Estimating Tree for Large-Scale RFID Estimation [J].

Zheng, Yuanqing ;

Li, Mo ;

Qian, Chen .

31ST INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS (ICDCS 2011), 2011, :37-46

[22] PET: Probabilistic Estimating Tree for Large-Scale RFID Estimation [J].

Zheng, Yuanqing ;

Li, Mo .

IEEE TRANSACTIONS ON MOBILE COMPUTING, 2012, 11 (11) :1763-1774

[23] Probabilistic and Distributed Control of a Large-Scale Swarm of Autonomous Agents [J].

Bandyopadhyay, Saptarshi ;

Chung, Soon-Jo ;

Hadaegh, Fred Y. .

IEEE TRANSACTIONS ON ROBOTICS, 2017, 33 (05) :1103-1123

[24] Probabilistic Diagnosis of Performance Faults in Large-Scale Parallel Applications [J].

Laguna, Ignacio ;

Ahn, Dong H. ;

de Supinski, Bronis R. ;

Bagchi, Saurabh ;

Gamblin, Todd .

PROCEEDINGS OF THE 21ST INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES (PACT'12), 2012, :213-222

[25] Tractable Learning and Inference for Large-Scale Probabilistic Boolean Networks [J].

Apostolopoulou, Ifigeneia ;

Marculescu, Diana .

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (09) :2720-2734

[26] A novel probabilistic generator for large-scale gene association networks [J].

Grimes, Tyler ;

Datta, Somnath .

PLOS ONE, 2021, 16 (11)

[27] A Large-Scale Study of Probabilistic Calibration in Neural Network Regression [J].

Dheur, Victor ;

Ben Taieb, Souhaib .

INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 202, 2023, 202

[28] Benders decomposition for the large-scale probabilistic set covering problem [J].

Liang, Jie ;

Yu, Cheng-Yang ;

Lv, Wei ;

Chen, Wei-Kun ;

Dai, Yu-Hong .

COMPUTERS & OPERATIONS RESEARCH, 2025, 177

[29] Scalable probabilistic PCA for large-scale genetic variation data [J].

Agrawal, Aman ;

Chiu, Alec M. ;

Le, Minh ;

Halperin, Eran ;

Sankararaman, Sriram .

PLOS GENETICS, 2020, 16 (05)

[30] Stabilizing Large-Scale Probabilistic Boolean Networks by Pinning Control [J].

Lin, Lin ;

Cao, Jinde ;

Lu, Jianquan ;

Zhong, Jie ;

Zhu, Shiyong .

IEEE TRANSACTIONS ON CYBERNETICS, 2022, 52 (12) :12929-12941

← 1 2 3 4 5 →