.png){kind=icon}
.png){kind=icon}
Return
A memory constrained bayesian optimization via robust online memory estimation
.png){kind=icon}
OA
AI
DOI：10.1007/s40747-025-02150-x.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Bayesian optimization (BO) is a memory-intensive algorithm that requires training and evaluating an expensive objective function. In contrast to previous works that use an offline memory estimation to make BO memory-efficient, we propose a robust and simple online memory estimation method that requires training a model only for the first two iterations of the first epoch. Our memory estimation method is then integrated with a simple, performance-based surrogate model of BO in a seamless (or in sync) mode that enforces memory efficiency even if it does not bypass a preset threshold. The online memory estimation method has been evaluated on two different datasets, showing that it is more accurate than the existing offline method ( $$2.19\times $$ for MNIST and $$3.51\times $$ for CIFAR datasets). Furthermore, compared to a memory-unaware baseline, the enhanced BO has no loss of accuracy and is $$11.31\times $$ memory-efficient for a simple CNN-based image classification, and $$5.03\times $$ memory efficient but $$9.27\times $$ slower for a more complex LSTM-based text classification (useful for a resource-constrained environment where delay is tolerable but memory is scarce), while it is $$2.6\times $$ memory efficient but $$1.23\times $$ slower on a pretrained ResNet50 model.
Keywords:
Online memory estimation
Bayesian optimization
Memory-efficient
Robust
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.6
Papers:
2.1K
Citations:
6.6K