PLAM: a Posit Logarithm-Approximate Multiplier
This work addresses hardware efficiency for deep neural network accelerators using posit numbers, representing an incremental improvement over existing posit multipliers.
The paper tackles the high complexity of posit multipliers in deep neural networks by proposing a Posit Logarithm-Approximate Multiplication (PLAM) scheme, which reduces area, power, and delay by up to 72.86%, 81.79%, and 17.01% respectively without accuracy loss.
The Posit Number System was introduced in 2017 as a replacement for floating-point numbers. Since then, the community has explored its application in Neural Network related tasks and produced some unit designs which are still far from being competitive with their floating-point counterparts. This paper proposes a Posit Logarithm-Approximate Multiplication (PLAM) scheme to significantly reduce the complexity of posit multipliers, the most power-hungry units within Deep Neural Network architectures. When comparing with state-of-the-art posit multipliers, experiments show that the proposed technique reduces the area, power, and delay of hardware multipliers up to 72.86%, 81.79%, and 17.01%, respectively, without accuracy degradation.