Saved in:
Bibliographic Details
Main Authors: Athanasiadis, Theodoros, Adriaensen, Steven, Müller, Samuel, Hutter, Frank
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.19733
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908507404500992
author Athanasiadis, Theodoros
Adriaensen, Steven
Müller, Samuel
Hutter, Frank
author_facet Athanasiadis, Theodoros
Adriaensen, Steven
Müller, Samuel
Hutter, Frank
contents The Adam optimizer remains one of the most widely used optimizers in deep learning, and effectively tuning its hyperparameters is key to optimizing performance. However, tuning can be tedious and costly. Freeze-thaw Bayesian Optimization (BO) is a recent promising approach for low-budget hyperparameter tuning, but is limited by generic surrogates without prior knowledge of how hyperparameters affect learning. We propose Adam-PFN, a new surrogate model for Freeze-thaw BO of Adam's hyperparameters, pre-trained on learning curves from TaskSet, together with a new learning curve augmentation method, CDF-augment, which artificially increases the number of available training examples. Our approach improves both learning curve extrapolation and accelerates hyperparameter optimization on TaskSet evaluation tasks, with strong performance on out-of-distribution (OOD) tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2508_19733
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Tune My Adam, Please!
Athanasiadis, Theodoros
Adriaensen, Steven
Müller, Samuel
Hutter, Frank
Machine Learning
The Adam optimizer remains one of the most widely used optimizers in deep learning, and effectively tuning its hyperparameters is key to optimizing performance. However, tuning can be tedious and costly. Freeze-thaw Bayesian Optimization (BO) is a recent promising approach for low-budget hyperparameter tuning, but is limited by generic surrogates without prior knowledge of how hyperparameters affect learning. We propose Adam-PFN, a new surrogate model for Freeze-thaw BO of Adam's hyperparameters, pre-trained on learning curves from TaskSet, together with a new learning curve augmentation method, CDF-augment, which artificially increases the number of available training examples. Our approach improves both learning curve extrapolation and accelerates hyperparameter optimization on TaskSet evaluation tasks, with strong performance on out-of-distribution (OOD) tasks.
title Tune My Adam, Please!
topic Machine Learning
url https://arxiv.org/abs/2508.19733