Novel Pruning of Dendritic Neuron Models for Improved System Implementation and Performance
- Resource Type
- Conference
- Authors
- Wen, Xiaohao; Zhou, MengChu; Luo, Xudong; Huang, Lukui; Wang, Ziyue
- Source
- 2021 IEEE International Conference on Systems, Man, and Cybernetics (SMC) Systems, Man, and Cybernetics (SMC), 2021 IEEE International Conference on. :1559-1564 Oct, 2021
- Subject
- Computing and Processing
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Training
Tensors
Computational modeling
Optimization methods
Dendrites (neurons)
Numerical models
Computational efficiency
Complex systems
Dendritic Neuron Model (DNM)
Machine learning
Neural network
Pruning
- Language
- ISSN
- 2577-1655
Pruning is widely used for neural network model compression. It removes redundant links from a weight tensor to lead to smaller and more efficient neural networks for system implementation. A compressed neural network can enable faster run and reduced computational cost in network training. In this paper, a novel pruning method is proposed for a dendritic neuron model (DNM). It calculates the significance of each DNM dendrite. The calculated significance is expressed numerically and a dendrite whose significance is lower than a pre-set threshold is removed. Experimental results verify that it obtains superior performance over the existing one in terms of both accuracy and computational efficiency.