FFGAF-SNN: The Forward-Forward Based Gradient Approximation Free Training Framework for Spiking Neural Networks

2025/07/31 by Changqing Xu, Xu, Changqing, Ziqiang Yang +10
Computer Science · Engineering · #Advanced Memory and Neural Computing #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Ferroelectric and Negative Capacitance Devices #Neural Networks and Reservoir Computing

paper · pdf · doi:10.48550/arxiv.2507.23643

openalex publication_date 2025/07/31 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Spiking Neural Networks (SNNs) offer a biologically plausible framework for energy-efficient neuromorphic computing. However, it is a challenge to train SNNs due to their non-differentiability, efficiently. Existing gradient approximation approaches frequently sacrifice accuracy and face deployment limitations on edge devices due to the substantial computational requirements of backpropagation. To address these challenges, we propose a Forward-Forward (FF) based gradient approximation-free training framework for Spiking Neural Networks, which treats spiking activations as black-box modules, thereby eliminating the need for gradient approximation while significantly reducing computational complexity. Furthermore, we introduce a class-aware complexity adaptation mechanism that dynamically optimizes the loss function based on inter-class difficulty metrics, enabling efficient allocation of network resources across different categories. Experimental results demonstrate that our proposed training framework achieves test accuracies of 99.58%, 92.13%, and 75.64% on the MNIST, Fashion-MNIST, and CIFAR-10 datasets, respectively, surpassing all existing FF-based SNN approaches. Additionally, our proposed method exhibits significant advantages in terms of memory access and computational power consumption.

Citations

• Backpropagation-free Spiking Neural Networks with the Forward-Forward Algorithm
• Training Spiking Neural Networks via Augmented Direct Feedback Alignment
• FFCL: Forward-Forward Net with Cortical Loops, Training and Inference on Edge Without Backpropagation
• The Trifecta: Three simple techniques for training deeper Forward-Forward networks
• SymBa: Symmetric Backpropagation-Free Contrastive Learning with Forward-Forward Algorithm for Optimizing Convergence
• Ultra-low Latency Adaptive Local Binary Spiking Neural Network with Accuracy Loss Estimator
• Robust and accelerated single-spike spiking neural network training with applicability to challenging temporal tasks
• Ultra-low Latency Spiking Neural Networks with Spatio-Temporal Compression and Synaptic Convolutional Block
• Going Deeper With Directly-Trained Larger Spiking Neural Networks
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Feed-Forward Optimization With Delayed Feedback for Neural Networks

Related

• Backpropagation-free Spiking Neural Networks with the Forward-Forward Algorithm
• Direct Training for Spiking Neural Networks: Faster, Larger, Better
• BioGrad: Biologically Plausible Gradient-Based Learning for Spiking Neural Networks
• Randomized Forward Mode Gradient for Spiking Neural Networks in Scientific Machine Learning
• Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-Based Optimization to Spiking Neural Networks