Compiler-based Processor Network Generation for Neural Networks on FPGAs

Plagwitz P, Hannig F, Teich J, Keszöcze O (2024)

Publication Language: English

Publication Type: Conference contribution, Original article

Publication year: 2024

Pages Range: 41-52

Conference Proceedings Title: Proceedings of the 27th Workshop on Methods and Description Languages for Modelling and Verification of Circuits and Systems (MBMV)

Event location: Kaiserslautern

ISBN: 978-3-8007-6267-5

Abstract

The fast-moving field of Neural Network (NN) research has brought breakthroughs in application domains like image or natural language processing. However, since new techniques are constantly emerging, fast prototyping of Neural Networks on edge devices like Field-Programmable Gate Arrays (FPGAs) is particularly appealing due to their reconfigurability. Existing compiler-based toolchains generating specialized hardware are rare and   restricted in their scope and coverage. In this paper, we propose a novel NN model-to-hardware flow that exploits the capabilities of Application-Specific Instruction Set Processors (ASIPs) and networks that can be built to implement NNs in a scalable and mixed hardware/software-reconfigurable way. Starting with an NN specification in PyTorch, an ASIP network generation toolchain is presented in which each instantiated core is assigned a portion of neurons to execute a given NN in a parallel and pipelined way while tailoring the instruction set of each core to the computations of the neurons. In addition, we propose multiple ML-specific instruction set extensions realized in hardware, including bit width-restricted ALU operations and activation functions with the circuits automatically generated from a given C/C++ implementation. Experimental results include a design space exploration of alternative implementations for the design objectives of FPGA resource usage,
latency, and throughput. Our approach is shown to achieve higher speed-ups than similar processor-based approaches while consuming lower amounts of resources.

Authors with CRIS profile

Related research project(s)

How to cite

APA:

Plagwitz, P., Hannig, F., Teich, J., & Keszöcze, O. (2024). Compiler-based Processor Network Generation for Neural Networks on FPGAs. In Proceedings of the 27th Workshop on Methods and Description Languages for Modelling and Verification of Circuits and Systems (MBMV) (pp. 41-52). Kaiserslautern, DE.

MLA:

Plagwitz, Patrick, et al. "Compiler-based Processor Network Generation for Neural Networks on FPGAs." Proceedings of the 27th Workshop on Methods and Description Languages for Modelling and Verification of Circuits and Systems (MBMV), Kaiserslautern 2024. 41-52.

BibTeX: Download