Final published version
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Chapter (peer-reviewed) › peer-review
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Chapter (peer-reviewed) › peer-review
}
TY - CHAP
T1 - Applications of Autonomous Learning Multi-model Systems
AU - Angelov, P.P.
AU - Gu, X.
PY - 2019
Y1 - 2019
N2 - In this chapter, the algorithm summaries of the autonomous learning multi-model systems of zero-order (ALMMo-0) and first-order (ALMMo-1) described in Chap. 8 are provided. Numerical examples based on well-known benchmark datasets are presented for evaluating the classification performance of the ALMMo-0 and ALMMo-1 systems. Real-world problems are also used for evaluating the performance of the ALMMo-1 system on regression. Numerical experiments and the comparison with the state-of-the-art approaches demonstrate that ALMMo systems can produce highly accurate classification and regression results on various problems after a very efficient training process. Furthermore, ALMMo systems can learn from streaming data on a sample-by-sample basis, self-evolve its system structure and self-update the meta-parameters continuously with newly observed data, which makes the ALMMo system a very attractive solution for various real world applications. The pseudo-code of the main procedure of the ALMMo-0 system and the MATLAB implementation are provided in appendices B.3 and C.3, and the corresponding pseudo-code and MATLAB implementation of ALMMo-1 systems are provided in appendices B.4 and C.4, respectively. © 2019, Springer Nature Switzerland AG.
AB - In this chapter, the algorithm summaries of the autonomous learning multi-model systems of zero-order (ALMMo-0) and first-order (ALMMo-1) described in Chap. 8 are provided. Numerical examples based on well-known benchmark datasets are presented for evaluating the classification performance of the ALMMo-0 and ALMMo-1 systems. Real-world problems are also used for evaluating the performance of the ALMMo-1 system on regression. Numerical experiments and the comparison with the state-of-the-art approaches demonstrate that ALMMo systems can produce highly accurate classification and regression results on various problems after a very efficient training process. Furthermore, ALMMo systems can learn from streaming data on a sample-by-sample basis, self-evolve its system structure and self-update the meta-parameters continuously with newly observed data, which makes the ALMMo system a very attractive solution for various real world applications. The pseudo-code of the main procedure of the ALMMo-0 system and the MATLAB implementation are provided in appendices B.3 and C.3, and the corresponding pseudo-code and MATLAB implementation of ALMMo-1 systems are provided in appendices B.4 and C.4, respectively. © 2019, Springer Nature Switzerland AG.
U2 - 10.1007/978-3-030-02384-3_12
DO - 10.1007/978-3-030-02384-3_12
M3 - Chapter (peer-reviewed)
SN - 9783030023836
VL - 800
T3 - Studies in Computational Intelligence
SP - 277
EP - 293
BT - Empirical Approach to Machine Learning
A2 - Angelov, Plamen
A2 - Gu, Xiaowei
PB - Springer-Verlag
ER -