xTern: Energy-Efficient Ternary Neural Network Inference on RISC-V-Based Edge Systems

TL;DR

xTern introduces a RISC-V ISA extension and optimized kernels that significantly improve energy efficiency and throughput for ternary neural network inference on edge systems, enabling more accurate and power-efficient AI deployment.

Contribution

The paper presents xTern, a novel ISA extension and kernel set that accelerates TNN inference on general-purpose cores, reducing power consumption and increasing throughput without significant hardware overhead.

Findings

Achieves 67% higher throughput than 2-bit equivalents.

Increases energy efficiency by 57.1%.

Enables deployment of more accurate TNNs with minimal hardware impact.

Abstract

Ternary neural networks (TNNs) offer a superior accuracy-energy trade-off compared to binary neural networks. However, until now, they have required specialized accelerators to realize their efficiency potential, which has hindered widespread adoption. To address this, we present xTern, a lightweight extension of the RISC-V instruction set architecture (ISA) targeted at accelerating TNN inference on general-purpose cores. To complement the ISA extension, we developed a set of optimized kernels leveraging xTern, achieving 67% higher throughput than their 2-bit equivalents. Power consumption is only marginally increased by 5.2%, resulting in an energy efficiency improvement by 57.1%. We demonstrate that the proposed xTern extension, integrated into an octa-core compute cluster, incurs a minimal silicon area overhead of 0.9% with no impact on timing. In end-to-end benchmarks, we demonstrate that xTern enables the deployment of TNNs achieving up to 1.6 percentage points higher CIFAR-10 classification accuracy than 2-bit networks at equal inference latency. Our results show that xTern enables RISC-V-based ultra-low-power edge AI platforms to benefit from the efficiency potential of TNNs.