Abstract
A variety of neural models, especially higher-order networks, are known to be computationally powerful for complex applications. While they have advantages over traditional multilayer perceptrons, the non-uniformity in their network structures and learning algorithms creates practical problems. Thus there is a need for a framework that unifies these various models. This paper presents a novel neuron model, called generalized multi-dendrite product (GMDP) unit. Multi-layer feedforward neural networks with GMDP units are shown to be capable of realizing higher-order neural networks. The standard backpropagation learning rule is extended to this neural network. Simulation results show that single layer GMDP networks provide an efficient model for solving general problems on function approximation and pattern classification.
| Original language | English |
|---|---|
| Pages | 107-112 |
| Number of pages | 6 |
| Publication status | Published - 1994 Dec 1 |
| Event | Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7) - Orlando, FL, USA Duration: 1994 Jun 27 → 1994 Jun 29 |
Other
| Other | Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7) |
|---|---|
| City | Orlando, FL, USA |
| Period | 94-06-27 → 94-06-29 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Software
Cite this
}
GMDP : a novel unified neuron model for multilayer feedforward neural networks. / Li, Sheng-Tun; Chen, Yiwei; Leiss, Ernst L.
1994. 107-112 Paper presented at Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7), Orlando, FL, USA, .Research output: Contribution to conference › Paper
TY - CONF
T1 - GMDP
T2 - a novel unified neuron model for multilayer feedforward neural networks
AU - Li, Sheng-Tun
AU - Chen, Yiwei
AU - Leiss, Ernst L.
PY - 1994/12/1
Y1 - 1994/12/1
N2 - A variety of neural models, especially higher-order networks, are known to be computationally powerful for complex applications. While they have advantages over traditional multilayer perceptrons, the non-uniformity in their network structures and learning algorithms creates practical problems. Thus there is a need for a framework that unifies these various models. This paper presents a novel neuron model, called generalized multi-dendrite product (GMDP) unit. Multi-layer feedforward neural networks with GMDP units are shown to be capable of realizing higher-order neural networks. The standard backpropagation learning rule is extended to this neural network. Simulation results show that single layer GMDP networks provide an efficient model for solving general problems on function approximation and pattern classification.
AB - A variety of neural models, especially higher-order networks, are known to be computationally powerful for complex applications. While they have advantages over traditional multilayer perceptrons, the non-uniformity in their network structures and learning algorithms creates practical problems. Thus there is a need for a framework that unifies these various models. This paper presents a novel neuron model, called generalized multi-dendrite product (GMDP) unit. Multi-layer feedforward neural networks with GMDP units are shown to be capable of realizing higher-order neural networks. The standard backpropagation learning rule is extended to this neural network. Simulation results show that single layer GMDP networks provide an efficient model for solving general problems on function approximation and pattern classification.
UR - http://www.scopus.com/inward/record.url?scp=0028730471&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0028730471&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:0028730471
SP - 107
EP - 112
ER -