What question did this study set out to answer?

The research investigates an alternative training method for spiking neural networks that avoids traditional backpropagation.

March 22, 2026Open Access

Backpropagation-free spiking neural networks with the forward–forward algorithm

Puntos clave

The research investigates an alternative training method for spiking neural networks that avoids traditional backpropagation.
Developed a framework based on the Forward-Forward (FF) algorithm for spiking neural networks.
Conducted evaluations on static datasets such as MNIST and spiking datasets like Neuro-MNIST and SHD.
Compared performance against existing backpropagation-trained SNNs and other FF-based models.
The FF-based SNN model demonstrated improved accuracy on static datasets compared to existing FF-based SNNs.
Achieved competitive performance on complex spiking tasks, outperforming many other SNN models.
Maintained a lighter architecture while achieving accuracy comparable to state-of-the-art backpropagation-trained SNNs.

Resumen

Spiking Neural Networks (SNNs) offer a biologically inspired computational paradigm that emulates neuronal activity through discrete spike-based processing. Despite their advantages, training SNNs with traditional backpropagation (BP) remains challenging due to computational inefficiencies and a lack of biological plausibility. This study explores the Forward–Forward (FF) algorithm as an alternative learning framework for SNNs. Unlike backpropagation, which relies on forward and backward passes, the FF algorithm employs two forward passes, enabling layer-wise localized learning, enhanced computational efficiency, and improved compatibility with neuromorphic hardware. We introduce an FF-based SNN training framework and evaluate its performance across both non-spiking (MNIST, Fashion-MNIST, Kuzushiji-MNIST) and spiking (Neuro-MNIST, SHD) datasets. Experimental results demonstrate that our model surpasses existing FF-based SNNs on evaluated static datasets with a much lighter architecture while achieving accuracy comparable to state-of-the-art backpropagation-trained SNNs. On more complex spiking tasks such as SHD, our approach outperforms other SNN models and remains competitive with leading backpropagation-trained SNNs. These findings highlight the FF algorithm’s potential to advance SNN training methodologies by addressing some key limitations of backpropagation.

Me gusta

Guardar

Ver artículo completo