We compute upper and lower bounds on the VC dimension and pseudo-dimension of feedforward neural networks composed of piecewise polynomial activation functions. We show that if the number of layers is fixed, then the VC dimension and pseudo-dimension grow as WlogW, where W is the number of parameters in the network. This result stands in opposition to the case where the number of layers is unbounded, in which case the VC dimension and pseudo-dimension grow as W2. We combine our results with recently established approximation error rates and determine error bounds for the problem of regression estimation by piecewise polynomial networks with unbounded weights.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1162/089976698300017016 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Association of Vitamin D Status with Immune Markers in a Cohort of Healthy Adults.
J Nutr
December 2024
Department of Nutrition, University of California, Davis, Davis, CA, United States; USDA Western Human Nutrition Research Center, University of California, Davis, Davis, CA, United States. Electronic address:
Background: Immune function is affected by vitamin D status but the optimal serum 25-hydroxy vitamin D [25(OH)D] level for immune function is not known.
Objectives: We hypothesized that 25(OH)D would be associated with markers of inflammation and immune activation.
Methods: We identified associations between 25(OH)D and immune markers from 361 healthy adults using polynomial regression.
A class of semiparametric models for bivariate survival data.
Lifetime Data Anal
December 2024
Statistics Department, Exact Sciences Institute, Federal University of Minas Gerais, Belo Horizonte, Brazil.
We propose a new class of bivariate survival models based on the family of Archimedean copulas with margins modeled by the Yang and Prentice (YP) model. The Ali-Mikhail-Haq (AMH), Clayton, Frank, Gumbel-Hougaard (GH), and Joe copulas are employed to accommodate the dependency among marginal distributions. Baseline distributions are modeled semiparametrically by the Piecewise Exponential (PE) distribution and the Bernstein polynomials (BP).
View Article and Find Full Text PDFDeveloping inverse motion planning technique for autonomous vehicles using integral nonlinear constraints.
Fundam Res
September 2024
Department of Civil Engineering, Toronto Metropolitan University, 350 Victoria Street, Toronto, ON M5B2K3, Canada.
The study considers issues of elaborating and validating a technique of autonomous vehicle motion planning based on sequential trajectory and speed optimization. This method includes components such as representing sought-for functions by finite elements (FE), vehicle kinematic model, sequential quadratic programming for nonlinear constrained optimization, and Gaussian N-point quadrature integration. The primary novelty consists of using the inverse approach for obtaining vehicle trajectory and speed.
View Article and Find Full Text PDFCell modeling using frequency modulation.
PLoS One
December 2024
Department of Electrical, Computer and Software Engineering, The University of Auckland, Auckland, New Zealand.
Computational models of the cell can be used to study the impact of drugs and assess pathological risks. Typically, these models are computationally demanding or challenging to implement in dedicated hardware for real-time emulation. A new Frequency Modulation (FM) model is proposed to address these limitations.
View Article and Find Full Text PDFApplication of CUSUM analysis in assessing learning curves in robot-assisted sacrocolpopexy performed by experienced gynecologist.
BMC Surg
December 2024
Department of Obstetrics and Gynecology, Kyung Hee University Hospital at Gangdong, School of Medicine, Kyung Hee University, 892, Dongnam-ro, Gangdong-gu, Seoul, 134-727, Korea.
Article Synopsis
- The study evaluated the learning curve for robotic-assisted sacrocolpopexy using CUSUM analysis to understand the surgeon's progress based on operation time, complications, and conversion rates to open surgery.
- A total of 50 surgeries performed by one experienced gynecologist were analyzed, identifying key breakpoints in the learning phases at approximately case 9 for proficiency and case 35 for competency.
- Results indicated that the surgeon faced no complications or need for conversion to open laparotomy throughout the study period, demonstrating successful learning and skill acquisition without any adverse outcomes.
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!