# SpikeFit: Towards Optimal Deployment of Spiking Networks on Neuromorphic Hardware

arXiv: https://arxiv.org/abs/2510.15542 · HTML: https://mlsanity.com/paper/2510.15542/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 47 references that are in the map.
- First posted on arXiv 17 Oct 2025.

## Builds on (most cited first, 24 of 24)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration](https://mlsanity.com/paper/2306.00978/) (2023)
- [Surrogate Gradient Learning in Spiking Neural Networks](https://mlsanity.com/paper/1901.09948/) (2019)
- [Pruning neural networks without any data by iteratively conserving synaptic flow](https://mlsanity.com/paper/2006.05467/) (2020)
- [HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precision](https://mlsanity.com/paper/1905.03696/) (2019)
- [The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits](https://mlsanity.com/paper/2402.17764/) (2024)
- [Training Spiking Neural Networks Using Lessons From Deep Learning](https://mlsanity.com/paper/2109.12894/) (2021)
- [Trained Ternary Quantization](https://mlsanity.com/paper/1612.01064/) (2016)
- [BitNet: Scaling 1-bit Transformers for Large Language Models](https://mlsanity.com/paper/2310.11453/) (2023)
- [Deep Rewiring: Training very sparse deep networks](https://mlsanity.com/paper/1711.05136/) (2017)
- [Convolutional Networks for Fast, Energy-Efficient Neuromorphic Computing](https://mlsanity.com/paper/1603.08270/) (2016)
- [Fisher Information and Natural Gradient Learning of Random Deep Networks](https://mlsanity.com/paper/1808.07172/) (2018)
- [Darwin3: A large-scale neuromorphic chip with a Novel ISA and On-Chip Learning](https://mlsanity.com/paper/2312.17582/) (2023)
- [QP-SNN: Quantized and Pruned Spiking Neural Networks](https://mlsanity.com/paper/2502.05905/) (2025)
- [End to End Binarized Neural Networks for Text Classification](https://mlsanity.com/paper/2010.05223/) (2020)
- [Towards Efficient Deep Spiking Neural Networks Construction with Spiking Activity based Pruning](https://mlsanity.com/paper/2406.01072/) (2024)
- [The fine line between dead neurons and sparsity in binarized spiking neural networks](https://mlsanity.com/paper/2201.11915/) (2022)
- [Towards Superior Quantization Accuracy: A Layer-sensitive Approach](https://mlsanity.com/paper/2503.06518/) (2025)
- [Hessian Aware Quantization of Spiking Neural Networks](https://mlsanity.com/paper/2104.14117/) (2021)
- [Low Precision Quantization-aware Training in Spiking Neural Networks with Differentiable Quantization Function](https://mlsanity.com/paper/2305.19295/) (2023)
- [A Power-Efficient Binary-Weight Spiking Neural Network Architecture for Real-Time Object Classification](https://mlsanity.com/paper/2003.06310/) (2020)
- [Rethinking Clustering for Robustness](https://mlsanity.com/paper/2006.07682/) (2020)
- [Prominent characteristics of recurrent neuronal networks are robust against low synaptic weight resolution](https://mlsanity.com/paper/2105.05002/) (2021)

## Built on by (most cited first, 2 of 2)

- [Quantization of Spiking Neural Networks Beyond Accuracy](https://mlsanity.com/paper/2604.14487/) (2026)
- [VQ4SNN: Vector Quantization for Memory-Efficient FPGA Spiking Neural Networks](https://mlsanity.com/paper/2606.20675/) (2026)

## Abstract (arXiv)

This paper introduces SpikeFit, a novel training method for Spiking Neural Networks (SNNs) that enables efficient inference on neuromorphic hardware, considering all its stringent requirements: the number of neurons and synapses that can fit on a single device, and lower bit-width representations (e.g., 4-bit, 8-bit). Unlike conventional compressing approaches that address only a subset of these requirements (limited numerical precision and limited number of neurons in the network), SpikeFit treats the allowed weights' discrete values themselves as learnable parameters co-optimized with the model, allowing for optimal Clusterization-Aware Training (CAT) of the model's weights at low precision (2-, 4-, or 8-bit) which results in higher network compression efficiency, as well as limiting the number of unique synaptic connections to a value required by neuromorphic processor. This joint optimization allows SpikeFit to find a discrete weight set aligned with hardware constraints, enabling the most complete deployment across a broader range of neuromorphic processors than existing methods of SNN compression support. Moreover, SpikeFit introduces a new hardware-friendly Fisher Spike Contribution (FSC) pruning method showing the state-of-the-art performance. We demonstrate that for spiking neural networks constrained to only four unique synaptic weight values (M = 4), our SpikeFit method not only outperforms state-of-the-art SNNs compression methods and conventional baselines combining extreme quantization schemes and clustering algorithms, but also meets a wider range of neuromorphic hardware requirements and provides the lowest energy use in experiments.