# Embedded FPGA Acceleration of Brain-Like Neural Networks: Online Learning to Scalable Inference

arXiv: https://arxiv.org/abs/2506.18530 · HTML: https://mlsanity.com/paper/2506.18530/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 23 references that are in the map.
- First posted on arXiv 23 Jun 2025.

## Builds on (most cited first, 4 of 4)

- [MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification](https://mlsanity.com/paper/2110.14795/) (2021)
- [StreamBrain: An HPC Framework for Brain-like Neural Networks on CPUs, GPUs and FPGAs](https://mlsanity.com/paper/2106.05373/) (2021)
- [Brain-like approaches to unsupervised learning of hidden representations -- a comparative study](https://mlsanity.com/paper/2005.03476/) (2020)
- [A Reconfigurable Stream-Based FPGA Accelerator for Bayesian Confidence Propagation Neural Networks](https://mlsanity.com/paper/2503.01561/) (2025)

## Built on by (most cited first, 1 of 1)

- [A Hardware-oriented Approach for Efficient Bayesian Inference Computation and Deployment](https://mlsanity.com/paper/2607.17855/) (2026)

## Abstract (arXiv)

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.