Maximizing the Directional Derivative
Vector Algebra: Method of Components
One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation
Gaussian Elimination: Problem Solving
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
Cluster Sampling Method
You might also read
Articles linked to this work by shared authors, journal, and citation graph.
Updated: Jun 20, 2026

Image Recognition and Parameter Analysis of Concrete Vibration State Based on Support Vector Machine
Published on: January 5, 2024
Yuta Ishikawa1, Ichiro Takeuchi, Ryohei Nakano
1Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya, Aichi 466-8555, Japan. ishikawa@goat.ics.nitech.ac.jp
This study introduces a novel multi-directional search method for Gaussian mixture model (GMM) estimation, improving upon existing variational Bayes (VB) techniques. The method effectively escapes local optima by utilizing saddle points for better GMM parameter estimation.
Area of Science:
Background:
Purpose of the Study:
Main Methods:
Main Results:
Conclusions: