arXiv Open Access 2025

Coding for Computation: Efficient Compression of Neural Networks for Reconfigurable Hardware

Hans Rosenberger Rodrigo Fischer Johanna S. Fröhlich Ali Bereyhi Ralf R. Müller
Lihat Sumber

Abstrak

As state of the art neural networks (NNs) continue to grow in size, their resource-efficient implementation becomes ever more important. In this paper, we introduce a compression scheme that reduces the number of computations required for NN inference on reconfigurable hardware such as FPGAs. This is achieved by combining pruning via regularized training, weight sharing and linear computation coding (LCC). Contrary to common NN compression techniques, where the objective is to reduce the memory used for storing the weights of the NNs, our approach is optimized to reduce the number of additions required for inference in a hardware-friendly manner. The proposed scheme achieves competitive performance for simple multilayer perceptrons, as well as for large scale deep NNs such as ResNet-34.

Topik & Kata Kunci

Penulis (5)

H

Hans Rosenberger

R

Rodrigo Fischer

J

Johanna S. Fröhlich

A

Ali Bereyhi

R

Ralf R. Müller

Format Sitasi

Rosenberger, H., Fischer, R., Fröhlich, J.S., Bereyhi, A., Müller, R.R. (2025). Coding for Computation: Efficient Compression of Neural Networks for Reconfigurable Hardware. https://arxiv.org/abs/2504.17403

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2025
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓