The paper introduces a diagnostic protocol to disentangle the contributions of parameter sharing and neighborhood access in graph neural networks (GNNs) for electricity load forecasting. By comparing a GNN to three reference models—per-node, shared-parameter without neighbors, and the same GNN without edges—the study evaluates the impact of graph structure across datasets from France, UK feeders, and client consumption series. Results show that parameter sharing often accounts for most of the performance gain, while edges contribute less and can be diminished by degree-preserving rewiring.
By Eloi Campagne (CB), Yvenn Amara-Ouali (LMO), Yannig Goude (LMO), Itai Zehavi (CB), Argyris Kalogeratos (CB)
arXiv:2608.20980v2 Announce Type: replace
Abstract: Graph neural networks (GNNs) are routinely employed for spatiotemporal forecasting, yet their performance across widely used benchmark datasets is...
By Kenneth Martin, Simon Heilig, Asja Fischer, Michel F. C. Haddad, Adam M. Sykulski, Moshe Eliasof
arXiv:2610.34611v1 Announce Type: new
Abstract: Spatio-temporal graph neural networks combine a node's history with information from graph neighbors. The central question is whether graph neighbors p...
By Saj Maru, Qiyu Han, Krishna Mamidala
arXiv:2608.20980v1 Announce Type: new
Abstract: Graph neural networks (GNNs) are routinely employed for short-range forecasting on multivariate time series with a spatial graph structure. Despite the...
By Kenneth Martin, Simon Heilig, Asja Fischer, Michel F. C. Haddad, Adam M. Sykulski, Moshe Eliasof
arXiv:1905. 11395v2 Announce Type: replace Abstract: Graph convolution network based approaches have been recently used to model region-wise relationships in region-level prediction problems in urban computing.
By Lingyu Zhang, Xu Geng, Zhiwei Qin, Hongjun Wang, Xiao Wang, Ying Zhang, Jian Liang, Guobin Wu, Xuan Song, Yunhai Wang
The paper presents an empirical benchmark of nine modern deep‑learning models for time‑series forecasting of smart‑meter energy consumption, evaluated on two publicly available datasets. It examines how historical input length, prediction horizon, and model architecture affect accuracy, finding that longer historical context improves performance up to a saturation point and that accuracy declines with longer horizons. The study also compares computational complexity, showing that lightweight architectures achieve similar performance to heavier models, and notes that model choice has limited impact across most demographic and household subgroups.
By Behnaz Kavoosighafi, Maria Eidenskog, Wiktoria Glad, Katerina Vrotsou
The paper presents an empirical benchmark of nine deep learning models for smart meter energy forecasting, evaluating them on two public datasets. It examines how historical input length, prediction horizon, and model architecture affect accuracy, finding that longer historical context improves performance up to a saturation point while accuracy declines with longer horizons. The study also compares computational cost, showing lightweight models achieve similar accuracy to heavier ones, and notes that model choice matters less across most population segments.
arXiv:2606. 19138v1 Announce Type: new Abstract: Neural Controlled Differential Equations (NCDE) provide a powerful continuous-time framework for forecasting time series, but standard graph-based extensions typically learn spatial structure purely from data, even in settings where a directed graph structure is known a priori.
By Michael Detzel, Gabriel Nobis, Kristiyan Blagov, Juri Schubert, Jackie Ma, Wojciech Samek
arXiv:2505. 13102v4 Announce Type: replace-cross Abstract: Unlike conventional "black-box" transformers with classical self-attention mechanism, we build a lightweight and interpretable transformer-like neural net by unrolling a mixed-graph-based optimization algorithm to forecast traffic with spatial and temporal dimensions.
By Ji Qi, Tam Thuc Do, Mingxiao Liu, Zhuoshi Pan, Yuzhe Li, Gene Cheung, H. Vicky Zhao
arXiv:2601. 19449v2 Announce Type: replace Abstract: Graph neural networks (GNNs) are widely believed to excel at node representation learning through trainable neighborhood aggregations.
By Celia Rubio-Madrigal, Rebekka Burkholz
arXiv:2501. 04339v2 Announce Type: replace-cross Abstract: We introduce the Deep Convolutional Interpreter for Time Series (DCIts), a deep-learning architecture for nonlinear multivariate time series that provides sample-specific, locally interpretable descriptions of the underlying interaction structure.
By Domjan Baric, Davor Horvatic
GNN‑VIEW is an interactive, browser‑based tool designed to aid the development and interpretation of graph‑based Earth‑system models. It visualizes the graph’s information pathways to help developers debug the structure before training, and after training it applies attribution‑based explainability to show which input features influence specific forecasts. The authors demonstrate its utility in building an ocean forecasting graph and in diagnosing poor precipitation predictions in a weather model during Hurricane Laura.
By Alessandro Lucatelli, Ferdinand Unterhuber, Katharina Krenzer, Mennatallah El-Assady, Benedikt Soja, Christina Humer, Joel Oskarsson