A
Amaury Lendasse
Researcher at University of Houston
Publications - 315
Citations - 7831
Amaury Lendasse is an academic researcher from University of Houston. The author has contributed to research in topics: Extreme learning machine & Feature selection. The author has an hindex of 39, co-authored 315 publications receiving 7167 citations. Previous affiliations of Amaury Lendasse include Ikerbasque & FedEx Institute of Technology.
Papers
More filters
Book ChapterDOI
A Novel ELM Ensemble for Time Series Prediction
TL;DR: This paper presents a novel methodology for time series prediction based on Extreme Learning Machines and an adaptive ensemble techniques that is tested successfully on the CIF 2016 competition datasets which are composed of 72 time series.
Posted Content
RMSE-ELM: Recursive Model based Selective Ensemble of Extreme Learning Machines for Robustness Improvement
TL;DR: RMSE-ELM as discussed by the authors is a two-layer recursive model, where the first layer trains lots of ELMs in different groups concurrently, and the second layer employs selective ensemble to pick out an optimal set of ELM in each group, which can be merged into a large group of ElMs called candidate pool.
14th IEEE international conference on Trust, Security, and Privacy in Computing and Communications (TrustCom), Helsinki, Finland, August 20-22, 2015
Luiza Sayfullina,Emil Eirola,Dmitry Komashinsky,Paolo Palumbo,Yoan Miche,Amaury Lendasse,Juha Karhunen +6 more
Book ChapterDOI
Mutual Information Based Initialization of Forward-Backward Search for Feature Selection in Regression Problems
TL;DR: New heuristics to find a more adequate starting point for the Forward-Backward Search algorithm are presented, based on the sorting of the variables using the Mutual Information criterion, and then performing parallel local searches.
Proceedings Article
EM-algorithm for Training of State-space Models with Application to Time Series Prediction
TL;DR: An improvement to the E step of the EM algorithm for nonlinear state-space models is presented and strategies for model structure selection when the EM-algorithm and state- space models are used for time series prediction are proposed.