Aug 2026· International Journal of Information Technology & Decision Making· 0 citations
TL;DR
Experimental results validate the effectiveness of the structure-adaptive threshold mechanism for low-power spiking graph learning and design an alternating soft-fusion-hard-grouping training strategy that decouples structure-aware threshold generation from pattern-specific threshold optimization.
Abstract
Spiking graph neural networks (SGNNs) have attracted considerable attention due to their high efficiency and low energy consumption in processing graph-structured data. In such networks, the firing threshold of neurons serves as a critical gating mechanism that governs spike sparsity, information flow, and energy consumption. However, existing threshold mechanisms are either fixed global constants or rely solely on coarse-grained statistics such as node degree. These approaches fundamentally fail to capture the rich local structural heterogeneity inherent in graph-structured data. In such data, nodes naturally assume distinct structural roles. Consequently, the resulting spike trains retain insufficient discriminative structural information. This deficiency compromises model capability and robustness against local structural distribution shifts. To address this issue, we propose a structure-adaptive threshold learning framework based on sparse representation. The framework learns a dictionary in which each atom captures a typical local subgraph pattern and is associated with a learnable threshold. Soft weights derived from sparse coding are leveraged to aggregate the atomic thresholds via weighted averaging, yielding a node-specific firing threshold. Furthermore, we design an alternating soft-fusion-hard-grouping training strategy that decouples structure-aware threshold generation from pattern-specific threshold optimization. Experimental results on multiple benchmark semi-supervised node classification tasks demonstrate that the proposed method significantly improves classification accuracy while preserving high spike sparsity. These results validate the effectiveness of the structure-adaptive threshold mechanism for low-power spiking graph learning.
This work proposes a community-aware sparse topology design framework for graph-based SNNs and challenges the prevailing assumption that network size or sparsity alone is sufficient, demonstrating that how sparse connections are organized - the graph topology - critically influences learning efficiency, accuracy, and energy consumption.
F. Motaghian, S. Nazari, J. P. Dominguez-Morales et al.· Scientific Reports· 0 citations
Across five graph benchmarks, the performance of SDRs is competitive with that of real-valued dense embeddings on node classification and link prediction, while the architecture exhibits portability, resilience to noise, and robustness to data corruption.
Understanding neuronal topology—how neurons are connected—is essential for uncovering neural computation principles and functional organization. However, accurately reconstructing such connectivity remains challenging due to the indirect nature of neural recordings and the complexity of network dynamics. As a first step towards this problem, a growing body of work has explored inferring monosynaptic connectivity directly from spike data. Among these, convolutional neural networks have shown promise when applied to spike-train cross-correlograms. Nevertheless, their ability to generalize across realistic experimental variability and the internal features that drive their predictions remain poorly understood. In this paper, we present a systematic benchmarking and diagnostic study of neural-network-based synaptic inference using simulations across a broad range of biophysical regimes. We show that connectivity classification and synaptic weight estimation, though often combined, rely on distinct internal representations and exhibit markedly different generalization behavior: robust connectivity models emphasize global structure in spike-train correlations, whereas weight estimation models are more sensitive to local signal amplitude and generalize less predictably. Importantly, we find that training on pooled, biologically grounded simulation data substantially improves robustness across parameter perturbations, outperforming models trained under narrow conditions. We further validate these findings in both simulated network data and an in vitro dataset from high‑density microelectrode array recordings with patch‑clamp‑verified ground‑truth connections. Models trained on diverse simulated circuits generalize effectively to novel network architectures and the experimental dataset. Together, these results demonstrate that incorporating biologically realistic diversity during training is critical for developing reliable machine-learning tools for large-scale synaptic inference from neural recordings.
Dynamic graph learning aims to capture evolving structural and semantic patterns in real-world systems, such as fraud detection and recommender systems. Due to the scarcity of labeled data in real-world dynamic graphs, recent studies have introduced generative or contrastive paradigms (e.g., masked graph autoencoders or graph contrastive learning) to generate task-agnostic graph embeddings. However, these methods typically rely on complex edge-level reconstruction objectives and tailored graph augmentation strategies. This incurs substantial computational overhead when scaling to large-scale dynamic graphs. In this paper, we propose SG-JEPA, a joint spiking embedding predictive architecture for large-scale dynamic graphs. In contrast to existing self-supervised methods, SG-JEPA partitions nodes into context and target sets along the temporal dimension to learn embeddings that are predictive of each other via additional spatial-temporal information. Furthermore, through encoding sequential inputs into coarse-to-fine spike count embeddings, spiking neurons enable SG-JEPA to adapt to the varying computational constraints of downstream tasks. Extensive experiments demonstrate that SG-JEPA achieves competitive or even superior performance over discriminative baselines on node classification, while effectively scaling to the dynamic graph with 13 million edges. SG-JEPA avoids the complex machinery (negative sampling, graph augmentations, edge-level reconstruction, etc.), resulting in superior training efficiency and memory scalability compared with prior self-supervised dynamic graph baselines.
Huizhe Zhang, Yuchang Zhu, Huazhen Zhong et al.· 0 citations
Deep neural networks often contain substantial parameter redundancy, resulting in unnecessary computational cost and energy consumption. This work presents a dynamically adaptive gating mechanism for learning layer-wise sparsity through differentiable masking. In the proposed framework, each network parameter is associated with a self-learning gate that controls its contribution during training. The gating function follows a progressive soft-to-hard transition in which the slope is gradually annealed, enabling the model to move smoothly from continuous parameter weighting to near-binary pruning decisions while simultaneously learning an adaptive threshold parameter. The framework is evaluated across multilayer perceptrons (MLPs), deep neural networks (DNNs), Tabular Transformer models, and benchmarked against $\mathrm{L}_{0}$ regularization and Variational Dropout on convolutional neural networks. Experimental results demonstrate substantial model compression without degrading predictive performance, and in some cases improving it. On MNIST, the gated MLP retains only 14% of weights (86% pruning) while maintaining 98% accuracy. The gated DNN and Tab Transformer similarly outperform their dense counterparts while retaining only 46% and 46.22% of parameters, respectively. Benchmarking further shows that the proposed adaptive gating achieves competitive or superior accuracy-sparsity trade-offs compared with $\mathbf{L}_{0}$ regularization and Variational Dropout. These results demonstrate that the proposed dynamically adaptive gating framework provides an efficient and interpretable pathway for sparsity learning, enabling high-performing lightweight neural network deployment.
Raunak Dev, Mydhily Sankar, Devaprabha Biju S et al.· 2026 International Conferenc...· 0 citations
Extensive experiments on static and neuromorphic benchmarks show that lower-bit BASC models match or outperform higher-bit baselines and retain this accuracy advantage after structured pruning, while further reducing model storage and synaptic operations.
Linliang Chen, Yan Zhong, Xin Liu et al.· 0 citations