Markov chain importance sampling for minibatches

This study investigates importance sampling under the scheme of minibatch stochastic gradient descent, under which the contributions are twofold. First, theoretically, we develop a neat tilting formula, which can be regarded as a general device for asymptotically optimal importance sampling. Second,...

Full description

Saved in:
Bibliographic Details
Published in:Machine learning 2024-02, Vol.113 (2), p.789-814
Main Authors: Fuh, Cheng-Der, Wang, Chuan-Ju, Pai, Chen-Hung
Format: Article
Language:English
Subjects:
Algorithms
Artificial Intelligence
Artificial neural networks
Computer Science
Control
Importance sampling
Machine Learning
Markov analysis
Markov chains
Mechatronics
Multilayer perceptrons
Natural Language Processing (NLP)
Normal distribution
Robotics
Simulation and Modeling
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study investigates importance sampling under the scheme of minibatch stochastic gradient descent, under which the contributions are twofold. First, theoretically, we develop a neat tilting formula, which can be regarded as a general device for asymptotically optimal importance sampling. Second, practically, guided by the formula, we present an effective algorithm for importance sampling which accounts for the effects of minibatches and leverages the Markovian property of the gradients between iterations. Experiments conducted on artificial data confirm that our algorithm consistently delivers superior performance in terms of variance reduction. Furthermore, experiments carried out on real-world data demonstrate that our method, when paired with relatively straightforward models like multilayer perceptron and convolutional neural networks, outperforms in terms of training loss and testing error.
ISSN:0885-6125
1573-0565
DOI:10.1007/s10994-023-06489-5