Relation between weight size and degree of over-fitting in neural network regression.

Neural Netw

Mie University, 1577 Kurima-Machiya-cho, Tsu 514-8507, Japan.

Published: January 2008

This paper investigates the relation between over-fitting and weight size in neural network regression. The over-fitting of a network to Gaussian noise is discussed. Using re-parametrization, a network function is represented as a bounded function g multiplied by a coefficient c. This is considered to bound the squared sum of the outputs of g at given inputs away from a positive constant delta(n), which restricts the weight size of a network and enables the probabilistic upper bound of the degree of over-fitting to be derived. This reveals that the order of the probabilistic upper bound can change depending on delta(n). By applying the bound to analyze the over-fitting behavior of one Gaussian unit, it is shown that the probability of obtaining an extremely small value for the width parameter in training is close to one when the sample size is large.

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neunet.2007.11.001DOI Listing

Publication Analysis

Top Keywords

weight size
12
degree over-fitting
8
neural network
8
network regression
8
probabilistic upper
8
upper bound
8
over-fitting
5
network
5
relation weight
4
size
4

Similar Publications

Maximal oxygen uptake (VOmax) in healthy subjects is primarily limited by systemic oxygen delivery. In chronic kidney disease (CKD), VOmax is potentially reduced by both central and peripheral factors. We aimed to investigate the effect on VOpeak of adding arm exercise to leg exercise.

View Article and Find Full Text PDF

Background: The WHO considers anemia in pregnancy a severe public health issue when prevalence surpasses 40%. In response, we conducted a systematic review and meta-analysis to examine anemia among pregnant women in Egypt, focusing on its prevalence, determinants, and associated complications.

Methods: We conducted a systematic literature search for studies published between January 1, 2010, and August 18, 2024, to identify studies from Egypt reporting on anemia in pregnant women, including its prevalence, associated determinants, and complications.

View Article and Find Full Text PDF

Background: Identifying markers or genes crucial for growth traits in Zhongwei goats is pivotal for breeding. Pinpointing genetic factors linked to body size gain enhances breeding efficiency and economic value. In this study, we used the MGISEQ-T7 platform to re-sequence 240 Zhongwei goats (133 male; 107 female) belonging to 5 metrics of growth traits at different growth stages (40 days and 6 months, here in after referred to as 40d and 6 m), namely, Body Weight (BW), Body Height (BH), Body Length (BL), Chest Circumference (CC), Tube Circumference (TC) were examined.

View Article and Find Full Text PDF

Role of Endoluminal Functional Lumen Imaging Probe in Bariatric Surgery and Bariatric Endoscopy.

Obes Surg

January 2025

Division of Gastroenterology and Hepatology, Department of Medicine, School of Medicine, Washington University, St. Louis, MO, 63130, USA.

Obesity is associated with significant morbidity, with an estimated prevalence of 42.4% in the USA. Treatment of severe obesity often involves Roux-en-Y gastric bypass and laparoscopic sleeve gastrectomy.

View Article and Find Full Text PDF

Enhancing biofuel pellet quality using torrefaction and co-pelletization of palm kernel shell and empty fruit bunch.

Environ Sci Pollut Res Int

January 2025

Faculty of Chemical & Process Engineering Technology, Universiti Malaysia Pahang Al-Sultan Abdullah, Lebuh Persiaran Tun Khalil Yaakob, 26300, Kuantan, Pahang, Malaysia.

Palm kernel shell (PKS) and empty fruit bunch (EFB) are potential biomass resources for producing solid biofuel for energy applications. However, raw EFB and PKS are not uniform in size and pose rotting behavior. Torrefaction and co-pelletization are both effective methods to improve their combustion and mechanical characteristics.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!