MAFIA: Machine Learning Acceleration on FPGAs for IoT Applications

TL;DR

MAFIA is a novel FPGA compilation tool optimized for IoT ML inference, outperforming existing HLS tools by 2.5x through native support for linear algebra and diverse ML models.

Contribution

Introduces MAFIA, a specialized FPGA compiler for small IoT devices that enhances performance by leveraging ML-specific properties.

Findings

MAFIA outperforms commercial HLS compiler variants by 2.5x on average.

Supports a wide range of ML algorithms and models.

Optimized for small form-factor FPGAs in IoT applications.

Abstract

Recent breakthroughs in ML have produced new classes of models that allow ML inference to run directly on milliwatt-powered IoT devices. On one hand, existing ML-to-FPGA compilers are designed for deep neural-networks on large FPGAs. On the other hand, general-purpose HLS tools fail to exploit properties specific to ML inference, thereby resulting in suboptimal performance. We propose MAFIA, a tool to compile ML inference on small form-factor FPGAs for IoT applications. MAFIA provides native support for linear algebra operations and can express a variety of ML algorithms, including state-of-the-art models. We show that MAFIA-generated programs outperform best-performing variant of a commercial HLS compiler by 2.5x on average.