Skip to content
Book Open access

Algorithm--Hardware Co-Design of Spiking Transformers for In-Memory Neuromorphic Edge Vision

Sep 2026 · Proceedings of the International Conference on Parallel Processing · pp. 436-445 · 0 citations · 14 references

Abstract

Reliable, ultra-low-power neuromorphic vision on edge devices requires attention mechanisms that combine accuracy with parallel, memory-efficient execution. Current Spiking Transformers suit neuromorphic data but retain the \(\mathcal {O}(N^2D)\) complexity of standard self-attention and weak temporal modeling, limiting their efficiency on in-memory edge hardware. We propose the Temporal Hadamard Transformer (THT), a spiking model co-designed with hardware for in-memory edge vision. Its Temporal Hadamard Attention (THA) combines a lightweight Temporal Processing Unit (TPU) for short-term temporal fusion with a binary Mask&Add operator. By replacing Query–Key–Value computation with accumulate-and-fire logic that maps efficiently onto memristor crossbars, THA reduces attention complexity to \(\mathcal {O}(ND)\) and enables highly parallel in-memory execution. On CIFAR10-DVS, THT achieves a peak Accuracy-to-Energy ratio of 157.06. THT consumes 0.51–2.19 mJ per inference. Its lowest-energy configuration uses up to 16 × lower energy than the compared models, while the largest configuration reaches 81.8% accuracy. We further design a detailed hardware deployment scheme and evaluate its robustness through non-ideal memristor crossbar simulations. Accounting for conductance drift and IR drop, the simulated deployment maintains 81.1% accuracy. Thus, THT integrates sparse algorithms with parallel hardware execution, providing a practical path toward real-time edge systems under realistic resource constraints and physical device non-idealities.

Read PDF

Similar papers

Open access Oct 2026

Beyond binary SNNs: a hardware-algorithm co-design and architectural evaluation for graded spike networks

While Spiking Neural Networks (SNNs) offer a promising path toward energy-efficient edge intelligence, conventional Binary representations often suffer from high inference latency and discretization errors. Multi-level (graded) spike models address these limitations by increasing information density per pulse, yet they...

Wen-Fei Song, A. Castagnetti, Pierre-Emmanuel Novac et al. · 0 citations

MorphAtt: A Neuromorphic Accelerator for Efficient Multi-Head Attention Processing in Spiking Vision Transformers

The proposed MorphAtt is a novel digital accelerator that expedites SViT inference through streamlined processing and offers better performance and efficiency trade-offs than state-of-the-art, thereby enabling highly energy-efficient vision-based AI systems at the edge.

Rachmad Vidya Wicaksana Putra, Amirhesam Jafari Rad, M. Shafique · 0 citations
Review Open access Sep 2026

Spiking Neural Network Chips for Low-Power Real-Time Edge Intelligence

This review first explains the signal path of an SNN chip through four connected modules: spike encoding, neuron computation, synaptic learning, and output decoding, and compares three implementation routes: digital many-core platforms, mixed-signal or sensor-integrated platforms, and memristive in-memory hardware.

Le-Yang Qin · 0 citations
#edge computing Review Open access Sep 2026

GPU and RISC-V acceleration for neuromorphic computing based on spiking neural networks: taxonomy, comparison, and open challenges

This survey reviews 71 studies published between 2015 and 2025 and organizes them into a taxonomy that classifies GPU works into simulation frameworks, training-acceleration techniques, large-scale and multi-GPU simulation, and application deployments, and RISC-V works into instruction-set extensions and tightly-couple...

Edris Zaman Farsa, Amirhossein Ilkhani, Marc Reichenbach et al. · 0 citations
#edge computing Open access Oct 2026

SpikeEdge: A Residual Spiking Neural Network Architecture With Edge-Based Data Encoding for Efficient Hardware Design

Deep spiking neural networks (SNNs) have gained significant attention in the current landscape of neural computing, primarily due to their remarkable energy efficiency and capability to perform complex computational tasks. This article presents the hardware implementation of a residual SNN (RSNN) architecture featuring...

Saras Mani Mishra, Souraja Kundu, Naorem Yaipharenba Meitei et al. · 0 citations
Open access Aug 2026

A hardware-implemented adaptive pruning neural network

An adaptive pruning convolutional neural network (APCNN) based on Si-based ferroelectric ambipolar field-effect transistor (Fe-AmFET) crossbar arrays is demonstrated, achieving an ultrahigh kernel pruning ratio with negligible accuracy degradation on the Fashion-MNIST dataset.

Yusen Tian, Penghao Chen, Ziyu Ming et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.