Abstract:Based on the network structure and training methods of extreme learning machines, extreme learning machine combining hidden-layer feature weighting and batch training (ELM-WB) is proposed to make full use of representation-level features for object images and human action videos classification. To solve the problem of insufficient fusion of multiple representation-level features in most classification methods, a double hidden layer structure in which the input layer and the second hidden layer are directly connected is designed. A loop training method of weighting coefficients and output weights is proposed based on the advantages of this structure. The proposed network structure and training method are combined to construct an extreme learning machine combining hidden-layer feature weighting (ELM-W), which can effectively fuse representation-level features to enhance the classification ability of representation-level features. On this basis, the principle of online sequential ELM (OS-ELM) is introduced to update the loop training formula of the two weights to reduce the memory consumption during the operation of the overall algorithm. ELM-WB is proposed by combining the loop training formula of two weight matrices with batch training. In order to test the feasibility of ELM-WB, experiments are conducted on Caltech 101, MSRC, UCF11 and UCF101 databases. Experimental results prove that the proposed ELM-WB can improve classification accuracy by fusing representation-level features. At the same time, ELM-WB can be used to perform classification tasks on databases of any size in a general-purpose computer without specific hardware.

Online Sequential Complex-Valued ELM for Noncircular Signals: Augmented Structures and Learning Algorithms

The augmented complex-valued extreme learning machine.

Multiple-kernel-learning-based Extreme Learning Machine for Classification Design

Augmented Online Sequential Quaternion Extreme Learning Machine

A Constructive Enhancement for Online Sequential Extreme Learning Machine.

Adaptive Online Sequential Extreme Learning Machine for Dynamic Modeling

An online sequential learning algorithm for regularized Extreme Learning Machine.

An Incremental Inverse-Free Augmented Complex-Valued Extreme Learning Machine

A Complex Hebbian-type Learning Algorithm with Application to Complex Adaptive ⅡR Filtering

Online sequential ELM algorithm and its improvement

Online Sequential Extreme Learning Machine Algorithms Based on Maximum Correntropy Citerion

Online sequential extreme learning machine with the increased classes

An Enhanced Online Sequential Extreme Learning Machine algorithm

A Novel Online Sequential Learning Algorithm for ELM Based on Optimal Control

An Augmented Echo State Network for Nonlinear Adaptive Filtering of Complex Noncircular Signals

Online Sequential ELM Algorithm with Forgetting Factor for Real Applications

Distributed Extreme Learning Machine for Nonlinear Learning over Network

An Adaptive Online Sequential Extreme Learning Machine for Real-Time Tidal Level Prediction

Voting base online sequential extreme learning machine for multi-class classification

Kernel Online Sequential ELM Algorithm with Sliding Window Subject to Time-Varying Environments

Extreme Learning Machine Combining Hidden-Layer Feature Weighting and Batch Training for Classification