Hebbian-LMS algorithm and its application to clustering
Abstract/Contents
- Abstract
- The Hebbian-LMS algorithm is a new unsupervised learning algorithm of an artificial neuron. It comes from the combination of two important concepts: Hebbian learning in neurobiology, and the Least Mean Square algorithm in signal processing. Hebbian learning is widely accepted as a basic theory of synaptic plasticity, which is essential for learning and memory in the human brain. The Least Mean Square algorithm is also widely used in many engineering areas, such as channel equalization or machine learning. The Hebbian-LMS algorithm is simpler and more biologically plausible than other training algorithms of artificial neural networks because the training of each artificial neuron takes place locally following Hebbian learning theory. Thanks to the simplicity, it is easy to apply the Hebbian-LMS algorithm to complicated artificial neural network structures, and the Hebbian-LMS clustering algorithm is introduced on multiple-layer artificial neural networks. It is shown that more artificial neurons and more layers improve the result of the Hebbian-LMS clustering algorithm. Furthermore, the Hebbian-LMS clustering algorithm is tested for various synthetic and real datasets to show that it is a decent clustering algorithm.
Description
Type of resource | text |
---|---|
Form | electronic; electronic resource; remote |
Extent | 1 online resource. |
Publication date | 2017 |
Issuance | monographic |
Language | English |
Creators/Contributors
Associated with | Kim, Youngsik |
---|---|
Associated with | Stanford University, Department of Electrical Engineering. |
Primary advisor | Poon, Ada Shuk Yan |
Primary advisor | Widrow, Bernard, 1929- |
Thesis advisor | Poon, Ada Shuk Yan |
Thesis advisor | Widrow, Bernard, 1929- |
Thesis advisor | Osgood, Brad |
Advisor | Osgood, Brad |
Subjects
Genre | Theses |
---|
Bibliographic information
Statement of responsibility | Youngsik Kim. |
---|---|
Note | Submitted to the Department of Electrical Engineering. |
Thesis | Thesis (Ph.D.)--Stanford University, 2017. |
Location | electronic resource |
Access conditions
- Copyright
- © 2017 by Youngsik Kim
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...