MINIMUM FIDELITY FOR RELIABLE ARCHITECTURE RANKING IN BAYESIAN NAS FOR OBJECT DETECTION.

Saved in:

Bibliographic Details
Title:	MINIMUM FIDELITY FOR RELIABLE ARCHITECTURE RANKING IN BAYESIAN NAS FOR OBJECT DETECTION.
Alternate Title:	Мінімальна точність для надійного рейтингу архітектури в баєсівському NAS для виявлення об’єктів.
Authors:	Kot, Anatoly¹ anatoly.kot@gmail.com
Source:	Electronics & Control Systems. 2026, Vol. 88 Issue 2, p54-60. 7p.
Subjects:	Neural architecture search, Rank correlation (Statistics), Mathematical optimization, Object recognition (Computer vision), Computer performance
Abstract:	This paper addresses the problem of reducing computational costs in Neural Architecture Search for object detection. The key question in low-fidelity approaches is: after what minimum number of epochs does the early training signal already provide acceptable architecture ranking? This paper presents a methodology for determining minimum fidelity based on out-of-sample rank correlation: trials are split into calibration (70%) and test (30%) sets, a composite proxy is built on the first set and evaluated on the second. An empirical study was conducted on 60 architectures trained for 25 epochs on an object detection dataset (6,772 images, 6 classes). The results show that a rank ensemble of three training metrics (val_loss, train_accuracy, val_accuracy) achieves Spearman ρ = 0.877 out-of-sample at epoch 3, providing 88% computational savings. A more complex 7-component metric underperforms the simple ensemble due to overfitting on a small sample (ρ_test = 0.70 vs. 0.88). The results are locally valid within the studied search space and the specific dataset. [ABSTRACT FROM AUTHOR]
	Copyright of Electronics & Control Systems is the property of National Aviation University and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	This paper addresses the problem of reducing computational costs in Neural Architecture Search for object detection. The key question in low-fidelity approaches is: after what minimum number of epochs does the early training signal already provide acceptable architecture ranking? This paper presents a methodology for determining minimum fidelity based on out-of-sample rank correlation: trials are split into calibration (70%) and test (30%) sets, a composite proxy is built on the first set and evaluated on the second. An empirical study was conducted on 60 architectures trained for 25 epochs on an object detection dataset (6,772 images, 6 classes). The results show that a rank ensemble of three training metrics (val_loss, train_accuracy, val_accuracy) achieves Spearman ρ = 0.877 out-of-sample at epoch 3, providing 88% computational savings. A more complex 7-component metric underperforms the simple ensemble due to overfitting on a small sample (ρ_test = 0.70 vs. 0.88). The results are locally valid within the studied search space and the specific dataset. [ABSTRACT FROM AUTHOR]
ISSN:	19905548
DOI:	10.18372/1990-5548.88.20966