IMPLY-based Approximate Full Adders for Efficient Arithmetic Operations in Image Processing and Machine Learning
Melanie Qiu, Caoyueshan Fan, Gulafshan, Salar Shakibhamedan, Fabian Seiler, Nima TaheriNejad
TL;DR
The paper tackles the energy and latency challenges in contemporary computing by leveraging memristor-based in-memory computing and approximate arithmetic. It introduces two serial IMPLY-based approximate full adders, SAPPI-1 and SAPPI-2, designed to preserve input states while reducing computation steps and energy in ripple-carry adders. Circuit-level simulations demonstrate substantial energy and step reductions (up to about 39–42% energy and 39–41% fewer steps) with competitive image-processing and MNIST-based ML performance, including PSNR maintenance for moderate approximations. The work demonstrates practical impact by enabling energy-efficient arithmetic cores that accelerate image processing and neural network inference with controlled accuracy loss.
Abstract
To overcome the performance limitations in modern computing, such as the power wall, emerging computing paradigms are gaining increasing importance. Approximate computing offers a promising solution by substantially enhancing energy efficiency and reducing latency, albeit with a trade-off in accuracy. Another emerging method is memristor-based In-Memory Computing (IMC) which has the potential to overcome the Von Neumann bottleneck. In this work, we combine these two approaches and propose two Serial APProximate IMPLY-based full adders (SAPPI). When embedded in a Ripple Carry Adder (RCA), our designs reduce the number of steps by 39%-41% and the energy consumption by 39%-42% compared to the exact algorithm. We evaluated our approach at the circuit level and compared it with State-of-the-Art (SoA) approximations where our adders improved the speed by up to 10% and the energy efficiency by up to 13%. We applied our designs in three common image processing applications where we achieved acceptable image quality with up to half of the RCA approximated. We performed a case study to demonstrate the applicability of our approximations in Machine Learning (ML) underscoring the potential gains in more complex scenarios. The proposed approach demonstrates energy savings of up to 296 mJ (21%) and a reduction of 1.3 billion (20%) computational steps when applied to Convolutional Neural Networks (CNNs) trained on the MNIST dataset while maintaining accuracy.