Embedded FPGA Acceleration of Brain-Like Neural Networks: Online Learning to Scalable Inference
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915355298889728 |
|---|---|
| author | Hafiz, Muhammad Ihsan Al Ravichandran, Naresh Lansner, Anders Herman, Pawel Podobas, Artur |
| author_facet | Hafiz, Muhammad Ihsan Al Ravichandran, Naresh Lansner, Anders Herman, Pawel Podobas, Artur |
| contents | Edge AI applications increasingly require models that can learn and adapt on-device with minimal energy budget. Traditional deep learning models, while powerful, are often overparameterized, energy-hungry, and dependent on cloud connectivity. Brain-Like Neural Networks (BLNNs), such as the Bayesian Confidence Propagation Neural Network (BCPNN), propose a neuromorphic alternative by mimicking cortical architecture and biologically-constrained learning. They offer sparse architectures with local learning rules and unsupervised/semi-supervised learning, making them well-suited for low-power edge intelligence. However, existing BCPNN implementations rely on GPUs or datacenter FPGAs, limiting their applicability to embedded systems. This work presents the first embedded FPGA accelerator for BCPNN on a Zynq UltraScale+ SoC using High-Level Synthesis. We implement both online learning and inference-only kernels with support for variable and mixed precision. Evaluated on MNIST, Pneumonia, and Breast Cancer datasets, our accelerator achieves up to 17.5x latency and 94% energy savings over ARM baselines, without sacrificing accuracy. This work enables practical neuromorphic computing on edge devices, bridging the gap between brain-like learning and real-world deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_18530 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Embedded FPGA Acceleration of Brain-Like Neural Networks: Online Learning to Scalable Inference Hafiz, Muhammad Ihsan Al Ravichandran, Naresh Lansner, Anders Herman, Pawel Podobas, Artur Hardware Architecture Artificial Intelligence Edge AI applications increasingly require models that can learn and adapt on-device with minimal energy budget. Traditional deep learning models, while powerful, are often overparameterized, energy-hungry, and dependent on cloud connectivity. Brain-Like Neural Networks (BLNNs), such as the Bayesian Confidence Propagation Neural Network (BCPNN), propose a neuromorphic alternative by mimicking cortical architecture and biologically-constrained learning. They offer sparse architectures with local learning rules and unsupervised/semi-supervised learning, making them well-suited for low-power edge intelligence. However, existing BCPNN implementations rely on GPUs or datacenter FPGAs, limiting their applicability to embedded systems. This work presents the first embedded FPGA accelerator for BCPNN on a Zynq UltraScale+ SoC using High-Level Synthesis. We implement both online learning and inference-only kernels with support for variable and mixed precision. Evaluated on MNIST, Pneumonia, and Breast Cancer datasets, our accelerator achieves up to 17.5x latency and 94% energy savings over ARM baselines, without sacrificing accuracy. This work enables practical neuromorphic computing on edge devices, bridging the gap between brain-like learning and real-world deployment. |
| title | Embedded FPGA Acceleration of Brain-Like Neural Networks: Online Learning to Scalable Inference |
| topic | Hardware Architecture Artificial Intelligence |
| url | https://arxiv.org/abs/2506.18530 |