Support Vector Machines Classification on Bendable RISC-V

2025-08-27

AI Summary

Machine learning deployment in flexible electronics faces critical challenges of large feature dimensions and high power consumption, hindering real-time, energy-efficient edge intelligence. Method: This work proposes a RISC-V-based heterogeneous architecture tailored for ultra-low-power edge AI, featuring a lightweight SVM accelerator with innovative support for one-versus-one (OvO) and one-versus-rest (OvR) multi-class strategies and configurable 4-/8-/16-bit weight quantization. The design tightly integrates a custom ML coprocessor with the open-source RISC-V instruction set to enable efficient vectorized computation and low-bitwidth inference. Contribution/Results: Evaluated on a flexible substrate, the system achieves a 21× average improvement in inference speed and energy efficiency over general-purpose processors. To our knowledge, this is the first open-source framework enabling real-time classification on bendable RISC-V systems, significantly advancing the development of low-cost, high-energy-efficiency flexible intelligent sensing devices.

Technology Category

• Machine Learning: Hardware-aware ML
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Learning & Optimization for CV

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Security and privacy of machine learning and AI applications
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

Flexible Electronics (FE) technology offers uniquecharacteristics in electronic manufacturing, providing ultra-low-cost, lightweight, and environmentally-friendly alternatives totraditional rigid electronics. These characteristics enable a rangeof applications that were previously constrained by the costand rigidity of conventional silicon technology. Machine learning (ML) is essential for enabling autonomous, real-time intelligenceon devices with smart sensing capabilities in everyday objects. However, the large feature sizes and high power consumption ofthe devices oppose a challenge in the realization of flexible ML applications. To address the above, we propose an open-source framework for developing ML co-processors for the Bendable RISC-V core. In addition, we present a custom ML accelerator architecture for Support Vector Machine (SVM), supporting both one-vs-one (OvO) and one-vs-rest (OvR) algorithms. Our ML accelerator adopts a generic, precision-scalable design, supporting 4-, 8-, and 16-bit weight representations. Experimental results demonstrate a 21x improvement in both inference execution time and energy efficiency, on average, highlighting its potential for low-power, flexible intelligence on the edge.

Problem

• Implementing SVM on flexible electronics with low power
• Overcoming large feature sizes in flexible ML applications
• Designing precision-scalable ML accelerator for Bendable RISC-V

Innovation

• Bendable RISC-V core ML co-processor framework
• Custom SVM accelerator supporting OvO and OvR
• Precision-scalable design with multiple bit representations