2β-softmax: A Hardware-Friendly Activation Function with Low Complexity and High Performance
- 6 October 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 353-354
Abstract
Softmax activation function is widely used in deep learning for multi-classification tasks. However, because of the exponentiation computation, its complexity is high in hardware implementation. Without adversely affecting mathematical characteristics and functions, we propose a new hardware-friendly softmax function: 2 β-softmax. The experiment based on MNIST dataset demonstrates the effectiveness of the proposed method. And compared with the traditional softmax function, its superiority lies in that the power consumption and area can be reduced by more than 50% in the hardware implementation.Keywords
Funding Information
- Nature
This publication has 3 references indexed in Scilit:
- Symmetric-Mapping LUT-Based Method and Architecture for Computing XY-Like FunctionsIEEE Transactions on Circuits and Systems I: Regular Papers, 2021
- A High-Speed and Low-Complexity Architecture for Softmax Function in Deep Learning2018
- A High Speed SoftMax VLSI Architecture Based on Basic-Split2018