Neural NetworksPub Date : 2025-04-24DOI: 10.1016/j.neunet.2025.107493
Yang Yang , Yuchao Gao , Hu Zhou , Jinran Wu , Shangce Gao , You-Gan Wang
{"title":"Multi-Granularity Autoformer for long-term deterministic and probabilistic power load forecasting","authors":"Yang Yang , Yuchao Gao , Hu Zhou , Jinran Wu , Shangce Gao , You-Gan Wang","doi":"10.1016/j.neunet.2025.107493","DOIUrl":"10.1016/j.neunet.2025.107493","url":null,"abstract":"<div><div>Long-term power load forecasting is critical for power system planning but is constrained by intricate temporal patterns. Transformer-based models emphasize modeling long- and short-term dependencies yet encounter limitations from complexity and parameter overhead. This paper introduces a novel Multi-Granularity Autoformer (MG-Autoformer) for long-term load forecasting. The model leverages a Multi-Granularity Auto-Correlation Attention Mechanism (MG-ACAM) to effectively capture fine-grained and coarse-grained temporal dependencies, enabling accurate modeling of short-term fluctuations and long-term trends. To enhance efficiency, a shared query–key (Q–K) mechanism is utilized to identify key temporal patterns across multiple resolutions and reduce model complexity. To address uncertainty in power load forecasting, the model incorporates a quantile loss function, enabling probabilistic predictions while quantifying uncertainty. Extensive experiments on benchmark datasets from Portugal, Australia, America, and ISO New England demonstrate the superior performance of the proposed MG-Autoformer in long-term power load point and probabilistic forecasting tasks.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107493"},"PeriodicalIF":6.0,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Neural NetworksPub Date : 2025-04-24DOI: 10.1016/j.neunet.2025.107494
Dong Zhang , Haoqian Jiang , Xiaoning Li , Guanyu Li , Bo Ning , Heng Chen
{"title":"Pair-wise or high-order? A self-adaptive graph framework for knowledge graph embedding","authors":"Dong Zhang , Haoqian Jiang , Xiaoning Li , Guanyu Li , Bo Ning , Heng Chen","doi":"10.1016/j.neunet.2025.107494","DOIUrl":"10.1016/j.neunet.2025.107494","url":null,"abstract":"<div><div>Knowledge graphs (KGs) depict entities as nodes and connections as edges, and they are extensively utilized in numerous artificial intelligence applications. However, knowledge graphs often suffer from incompleteness, which seriously affects downstream applications. Knowledge graph embedding (KGE) technology tackles this challenge by encoding entities and relations as vectors, allowing for inference and computations using these representations. Graph convolutional networks (GCNs) are essential knowledge graph embedding technology models. GCN methods capture the topological structure features of the KG by calculating the pair-wise relationships between entities and neighboring nodes. Although GCN models have achieved excellent performance, there are still three main challenges: (1) effectively solving the over-smoothing problem in multi-layer GCN models for knowledge graph representation learning, (2) obtaining high-order information between entities and neighboring nodes beyond the pair-wise relationships, and (3) effectively integrating pair-wise and high-order features of entities. To address these challenges, we proposed an adaptive graph convolutional network model called PHGCN (Pair-wise and High-Order Graph Convolutional Network), which can simultaneously integrate pair-wise and high-order features. PHGCN tackles the three challenges mentioned above in the following ways. (1) We propose a layer-aware GCN to overcome the over-smoothing problem while aggregating pair-wise relationships. (2) We employ simplicial complex neural networks to extract high-order topological features from knowledge graphs. (3) We introduce a self-adaptive aggregation mechanism that effectively integrates pair-wise and high-order features. Our experiments on four benchmark datasets showed that PHGCN outperforms existing methods, achieving state-of-the-art results. The performance improvement from using a simplicial complex neural network to extract high-order features is significant. On the FB15k-237 dataset, PHGCN achieved a 1.5% improvement, while on the WN18RR dataset, it improved by 6.1%.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107494"},"PeriodicalIF":6.0,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873744","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Neural NetworksPub Date : 2025-04-23DOI: 10.1016/j.neunet.2025.107479
Rui Gao, Weiwei Liu
{"title":"Red alarm: Controllable backdoor attack in continual learning","authors":"Rui Gao, Weiwei Liu","doi":"10.1016/j.neunet.2025.107479","DOIUrl":"10.1016/j.neunet.2025.107479","url":null,"abstract":"<div><div>Continual learning (CL) studies the problem of learning a single model from a sequence of disjoint tasks. The main challenge is to learn without catastrophic forgetting, a scenario in which the model’s performance on previous tasks degrades significantly as new tasks are added. However, few works focus on the security challenge in the CL setting. In this paper, we focus on the backdoor attack in the CL setting. Specifically, we provide the threat model and explore what attackers in a CL setting will face. Based on these findings, we propose a controllable backdoor attack mechanism in continual learning (CBACL). Experimental results on the Split Cifar and Tiny Imagenet datasets confirm the advantages of our proposed mechanism.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107479"},"PeriodicalIF":6.0,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143877169","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Neural NetworksPub Date : 2025-04-22DOI: 10.1016/j.neunet.2025.107511
Lincong Pan , Kun Wang , Yongzhi Huang , Xinwei Sun , Jiayuan Meng , Weibo Yi , Minpeng Xu , Tzyy-Ping Jung , Dong Ming
{"title":"Enhancing motor imagery EEG classification with a Riemannian geometry-based spatial filtering (RSF) method","authors":"Lincong Pan , Kun Wang , Yongzhi Huang , Xinwei Sun , Jiayuan Meng , Weibo Yi , Minpeng Xu , Tzyy-Ping Jung , Dong Ming","doi":"10.1016/j.neunet.2025.107511","DOIUrl":"10.1016/j.neunet.2025.107511","url":null,"abstract":"<div><div>Motor imagery (MI) refers to the mental simulation of movements without physical execution, and it can be captured using electroencephalography (EEG). This area has garnered significant research interest due to its substantial potential in brain-computer interface (BCI) applications, especially for individuals with physical disabilities. However, accurate classification of MI EEG signals remains a major challenge due to their non-stationary nature, low signal-to-noise ratio, and sensitivity to both external and physiological noise. Traditional classification methods, such as common spatial pattern (CSP), often assume that the data is stationary and Gaussian, which limits their applicability in real-world scenarios where these assumptions do not hold. These challenges highlight the need for more robust methods to improve classification accuracy in MI-BCI systems. To address these issues, this study introduces a Riemannian geometry-based spatial filtering (RSF) method that projects EEG signals into a lower-dimensional subspace, maximizing the Riemannian distance between covariance matrices from different classes. By leveraging the inherent geometric properties of EEG data, RSF enhances the discriminative power of the features while maintaining robustness against noise. The performance of RSF was evaluated in combination with ten commonly used MI decoding algorithms, including CSP with linear discriminant analysis (CSP-LDA), Filter Bank CSP (FBCSP), Minimum Distance to Riemannian Mean (MDM), Tangent Space Mapping (TSM), EEGNet, ShallowConvNet (sCNN), DeepConvNet (dCNN), FBCNet, Graph-CSPNet, and LMDA-Net, using six publicly available MI-BCI datasets. The results demonstrate that RSF significantly improves classification accuracy and reduces computational time, particularly for deep learning models with high computational complexity. These findings underscore the potential of RSF as an effective spatial filtering approach for MI EEG classification, providing new insights and opportunities for the development of robust MI-BCI systems. The code for this research is available at <span><span>https://github.com/PLC-TJU/RSF</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107511"},"PeriodicalIF":6.0,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143877170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Neural NetworksPub Date : 2025-04-22DOI: 10.1016/j.neunet.2025.107465
Johannes Pöppelbaum, Andreas Schwung
{"title":"Time series compression using quaternion valued neural networks and quaternion backpropagation","authors":"Johannes Pöppelbaum, Andreas Schwung","doi":"10.1016/j.neunet.2025.107465","DOIUrl":"10.1016/j.neunet.2025.107465","url":null,"abstract":"<div><div>We propose a novel quaternionic time series compression methodology where we divide a long time series into segments of data, extract the min, max, mean and standard deviation of these chunks as representative features and encapsulate them in a quaternion, yielding a quaternion valued time series. This time series is processed using quaternion valued neural network layers, where we aim to preserve the relation between these features through the usage of the Hamilton product. To train this quaternion neural network, we derive quaternion backpropagation employing the GHR calculus, which is required for a valid product and chain rule in quaternion space. Furthermore, we investigate the connection between the derived update rules and automatic differentiation.</div><div>We apply our proposed compression method on the Tennessee Eastman Dataset, where we perform fault classification using the compressed data in two settings: a fully supervised one and in a semi supervised, contrastive learning setting. Both times, we were able to outperform real valued counterparts as well as two baseline models: one with the uncompressed time series as the input and the other with a regular downsampling using the mean. Further, we could improve the classification benchmark set by SimCLR-TS from 81.43% to 83.90%.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107465"},"PeriodicalIF":6.0,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Neural NetworksPub Date : 2025-04-22DOI: 10.1016/j.neunet.2025.107481
Lexiang Hu , Yikang Li , Zhouchen Lin
{"title":"Symmetry discovery for different data types","authors":"Lexiang Hu , Yikang Li , Zhouchen Lin","doi":"10.1016/j.neunet.2025.107481","DOIUrl":"10.1016/j.neunet.2025.107481","url":null,"abstract":"<div><div>Equivariant neural networks incorporate symmetries into their architecture, achieving higher generalization performance. However, constructing equivariant neural networks typically requires prior knowledge of data types and symmetries, which is difficult to achieve in most tasks. In this paper, we propose LieSD, a method for discovering symmetries via trained neural networks which approximate the input–output mappings of the tasks. It characterizes equivariance and invariance (a special case of equivariance) of continuous groups using Lie algebra and directly solves the Lie algebra space through the inputs, outputs, and gradients of the trained neural network. Then, we extend the method to make it applicable to multi-channel data and tensor data, respectively. We validate the performance of LieSD on tasks with symmetries such as the two-body problem, the moment of inertia matrix prediction, top quark tagging, and rotated MNIST. Compared with the baseline, LieSD can accurately determine the number of Lie algebra bases without the need for expensive group sampling. Furthermore, LieSD can perform well on non-uniform datasets, whereas methods based on GANs fail. Code and data are available at <span><span>https://github.com/hulx2002/LieSD</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107481"},"PeriodicalIF":6.0,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Neural NetworksPub Date : 2025-04-21DOI: 10.1016/j.neunet.2025.107455
Liu Yang, Siting Liu, Stanley J. Osher
{"title":"Fine-tune language models as multi-modal differential equation solvers","authors":"Liu Yang, Siting Liu, Stanley J. Osher","doi":"10.1016/j.neunet.2025.107455","DOIUrl":"10.1016/j.neunet.2025.107455","url":null,"abstract":"<div><div>In the growing domain of scientific machine learning, in-context operator learning has shown notable potential in building foundation models, as in this framework the model is trained to learn operators and solve differential equations using prompted data, during the inference stage without weight updates. However, the current model’s overdependence on function data overlooks the invaluable human insight into the operator. To address this, we present a transformation of in-context operator learning into a multi-modal paradigm. In particular, we take inspiration from the recent success of large language models, and propose using “captions” to integrate human knowledge about the operator, expressed through natural language descriptions and equations. Also, we introduce a novel approach to train a language-model-like architecture, or directly fine-tune existing language models, for in-context operator learning. We beat the baseline on single-modal learning tasks, and also demonstrated the effectiveness of multi-modal learning in enhancing performance and reducing function data requirements. The proposed method not only significantly enhanced the development of the in-context operator learning paradigm, but also created a new path for the application of language models.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107455"},"PeriodicalIF":6.0,"publicationDate":"2025-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"SpikeCLIP: A contrastive language–image pretrained spiking neural network","authors":"Changze Lv , Tianlong Li , Wenhao Liu , Yufei Gu , Jianhan Xu , Cenyuan Zhang , Muling Wu , Xiaoqing Zheng , Xuanjing Huang","doi":"10.1016/j.neunet.2025.107475","DOIUrl":"10.1016/j.neunet.2025.107475","url":null,"abstract":"<div><div>Spiking Neural Networks (SNNs) have emerged as a promising alternative to conventional Artificial Neural Networks (ANNs), demonstrating comparable performance in both visual and linguistic tasks while offering the advantage of improved energy efficiency. Despite these advancements, the integration of linguistic and visual features into a unified representation through spike trains poses a significant challenge, and the application of SNNs to multimodal scenarios remains largely unexplored. This paper presents SpikeCLIP, a novel framework designed to bridge the modality gap in spike-based computation. Our approach employs a two-step recipe: an “alignment pre-training” to align features across modalities, followed by a “dual-loss fine-tuning” to refine the model’s performance. Extensive experiments reveal that SNNs achieve results on par with ANNs while substantially reducing energy consumption across various datasets commonly used for multimodal model evaluation. Furthermore, SpikeCLIP maintains robust image classification capabilities, even when dealing with classes that fall outside predefined categories. This study marks a significant advancement in the development of energy-efficient and biologically plausible multimodal learning systems.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107475"},"PeriodicalIF":6.0,"publicationDate":"2025-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}