Details

Autor(en) / Beteiligte

• Zhang, Tingting
• Niu, Zijing
• Jiang, Honglan
• Cockburn, Bruce F.
• Liu, Leibo
• Han, Jie

Titel

Logarithmic Floating-Point Multipliers for Efficient Neural Network Training

Ist Teil von

• Design and Applications of Emerging Computer Systems, p.567-587

Ort / Verlag

Cham: Springer Nature Switzerland

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Floating-point (FP) arithmetic computation is favored for training neural networks (NNs) due to its wide numerical range. The computation-intensive training process requires a tremendous amount of multiplication, which poses a challenge to deploying NN architectures on resource-constrained devices. This chapter presents hardware-efficient logarithmic FP multipliers (LFPMs) for NN training. By using piecewise approximations in different configurations over the applicable domains of the logarithm and anti-logarithm functions, we obtain LFPMs with various characteristics in accuracy and hardware. The benchmark NN applications are considered for evaluation.