Approximate Softmax Architecture for Energy-Efficient Deep Neural Networks
International Journal of Science, Strategic Management and Technology · 25.03.2026 · 02 (03) · 1-9
DOI: 10.55041/ijsmt.v2i3.262 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Science, Strategic Management and Technology, online erschienen am 25.03.2026. Verfasst von Athul Krishna R und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Softmax is commonly used in neural network- based classification systems to convert output values into probabilities, but its conventional implementation involves complex operations such as exponentials and division, which are inefficient for FPGA and VLSI-based hardware due to high area, power, and latency requirements. This project presents a hardware- efficient approximate softmax architecture designed for low-power FPGA systems using a simplified Top-1 approximation approach. The proposed design identifies the dominant output class using comparator-based logic and fixed- point arithmetic, thereby eliminating computationally expensive operations while maintaining correct decision-making. The approximate softmax module is implemented using synthesizable SystemVerilog and validated through simulation and synthesis using Xilinx Vivado, with Python used only for basic numerical verification. The developed design is hardware-ready and suitable for integration into FPGA-based neural network accelerators and embedded VLSI systems, with scope for future hardware implementation.
Zitieren
APA
R, A. K., J, D. S., A, F. M., & S, B. (2026). Approximate Softmax Architecture for Energy-Efficient Deep Neural Networks. International Journal of Science, Strategic Management and Technology, 02(03), 1-9. https://doi.org/10.55041/ijsmt.v2i3.262
Vancouver
R AK, J DS, A FM, S B. Approximate Softmax Architecture for Energy-Efficient Deep Neural Networks. International Journal of Science, Strategic Management and Technology. 2026;02(03):1-9. doi:10.55041/ijsmt.v2i3.262
BibTeX
@article{R2026,
author = {R, Athul Krishna and J, Deepak S and A, Fayas Mohamed and S, Baskar},
title = {Approximate Softmax Architecture for Energy-Efficient Deep Neural Networks},
journal = {International Journal of Science, Strategic Management and Technology},
year = {2026},
volume = {02},
number = {03},
pages = {1--9},
doi = {10.55041/ijsmt.v2i3.262}
}
EUVIMED