Efficient Traffic Prediction at Scale: A Systematic Study of STGCN Architectural Depth

• URL: http://arxiv.org/abs/2606.09539v1
• Date: Mon, 08 Jun 2026 14:23:56 GMT
• Title: Efficient Traffic Prediction at Scale: A Systematic Study of STGCN Architectural Depth
• Abstract summary: Single-block architecture achieves optimal performance for short-term prediction (10 mins) on three of four datasets.<n>The 2-block variant incurs 61% higher CPU inference latency and 37% lower throughput relative to 1-block.<n>The 3-block architecture offers no favourable tradeoff, more than doubling computational cost for $$0.5% relative improvement.
• Score: 9.291555165680863
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Spatio-temporal graph neural networks (STGNNs) have become the dominant approach for traffic prediction, yet their computational requirements pose challenges for practical deployment in intelligent transportation systems (ITS). While recent work has proposed efficient alternatives to STGNNs, a fundamental question remains unexplored: are these architectures themselves over-parameterised? We examine this question using the Spatio-Temporal Graph Convolutional Network (STGCN), one of the most widely adopted models in this domain. Through systematic experiments across four diverse traffic datasets, we compare 1-block, 2-block (standard), and 3-block STGCN variants. Our findings reveal that the single-block architecture achieves optimal performance for short-term prediction (10 mins) on three of four datasets, while incurring only marginal degradation ($\leq$1.8% relative error) at longer horizons. Crucially, the 2-block variant incurs 61% higher CPU inference latency and 37% lower throughput relative to 1-block -- substantial overhead for resource-constrained ITS deployment. The 3-block architecture offers no favourable tradeoff, more than doubling computational cost for $<$0.5% relative improvement. These results suggest that the default 2-block STGCN may be over-parameterised for many applications, with implications for both practitioners deploying traffic prediction systems and researchers benchmarking efficiency-focused methods.

Related papers

• Model-Free Surrogate-Assisted Neural Architecture Search for Evolving Variable-Length Dense Blocks [2.473241744699909]
  We propose a Model-Free Surrogate PSO Network (MFSPNet) for evolving convolutional neural network architectures.<n>MFSPNet integrates a lightweight model-free surrogate predictor within a particle swarm optimization (PSO) framework, eliminating the need for pre-trained surrogate models.<n>Under a consistent training protocol with ten independent runs, the proposed method attains error rates of 3.91%, 17.68%, and 1.91% on CIFAR-10, CIFAR-100, and SVHN, respectively.
  arXiv  Detail & Related papers  (2026-09-02T11:32:07Z)
• Leveraging Structural Constraints for Diffusion-based Neural TSP Solvers [3.588053519843615]
  State-ofthe-art approaches like FT2T combine fast consistency-based prediction with gradient-based inference time refinement.<n>We introduce Projected Consistency Inference ( PCI), a plug-and-play, retraining-free alternative that replaces gradient refinement with structure-aware projections.<n> PCI achieves an average optimality gap (OG) of 0.17% on the Euclidean Traveling Salesman Problem with 500 cities, and 0.31% on the Euclidean Traveling Salesman Problem with 1000 cities, outperforming FT2T best settings.
  arXiv  Detail & Related papers  (2026-06-08T11:08:36Z)
• QNAS: A Neural Architecture Search Framework for Accurate and Efficient Quantum Neural Networks [2.5435687567731926]
  We introduce QNAS, a neural architecture search framework that unifies aware hardware evaluation, multi objective optimization, and cutting overhead awareness for hybrid quantum classical neural networks (HQNNs)<n>QNAS evaluates candidate HQNNs under a few epochs of training and finds tradeoffs between accuracy, efficiency, and cutting overhead.<n>Across MNIST, Fashion-MNIST, and Iris benchmarks, we observe that embedding type and CNOT mode selection significantly impact both accuracy and efficiency.
  arXiv  Detail & Related papers  (2026-04-08T12:34:01Z)
• WarPGNN: A Parametric Thermal Warpage Analysis Framework with Physics-aware Graph Neural Network [1.4476963891735688]
  We present WarPGNN, an ef-ficient and accurate parametric thermal warpage analysis framework powered by Graph Neural Networks (GNNs)<n>Our method achieves more than 205.91x speedup compared with the 2-D efficient FEM-based method and over 119766.64x acceleration with 3-D FEM method COMSOL.<n>WarPGNN demonstrates remarkable transferability on unseen datasets with up to 3.69% normalized RMSE and similar runtime.
  arXiv  Detail & Related papers  (2026-03-19T07:40:57Z)
• Geminet: Learning the Duality-based Iterative Process for Lightweight Traffic Engineering in Changing Topologies [53.38648279089736]
  Geminet is a lightweight and scalable ML-based TE framework that can handle changing topologies.<n>Its neural network size is only 0.04% to 7% of existing schemes.<n>When trained on large-scale topologies, Geminet consumes under 10 GiB of memory, more than eight times less than the 80-plus GiB required by HARP.
  arXiv  Detail & Related papers  (2025-06-30T09:09:50Z)
• FusionLLM: A Decentralized LLM Training System on Geo-distributed GPUs with Adaptive Compression [55.992528247880685]
  Decentralized training faces significant challenges regarding system design and efficiency. We present FusionLLM, a decentralized training system designed and implemented for training large deep neural networks (DNNs) We show that our system and method can achieve 1.45 - 9.39x speedup compared to baseline methods while ensuring convergence.
  arXiv  Detail & Related papers  (2024-10-16T16:13:19Z)
• Correlating sparse sensing for large-scale traffic speed estimation: A Laplacian-enhanced low-rank tensor kriging approach [76.45949280328838]
  We propose a Laplacian enhanced low-rank tensor (LETC) framework featuring both lowrankness and multi-temporal correlations for large-scale traffic speed kriging. We then design an efficient solution algorithm via several effective numeric techniques to scale up the proposed model to network-wide kriging.
  arXiv  Detail & Related papers  (2022-10-21T07:25:57Z)
• Binary Early-Exit Network for Adaptive Inference on Low-Resource Devices [3.591566487849146]
  Binary neural networks (BNNs) tackle the issue with extreme compression and speed-up gains compared to real-valued models. We propose a simple but effective method to accelerate inference through unifying BNNs with an early-exiting strategy. Our approach allows simple instances to exit early based on a decision threshold and utilizes output layers added to different intermediate layers to avoid executing the entire binary model.
  arXiv  Detail & Related papers  (2022-06-17T22:11:11Z)
• Learning to Solve Combinatorial Graph Partitioning Problems via Efficient Exploration [72.15369769265398]
  Experimentally, ECORD achieves a new SOTA for RL algorithms on the Maximum Cut problem. Compared to the nearest competitor, ECORD reduces the optimality gap by up to 73%.
  arXiv  Detail & Related papers  (2022-05-27T17:13:10Z)
• ZARTS: On Zero-order Optimization for Neural Architecture Search [94.41017048659664]
  Differentiable architecture search (DARTS) has been a popular one-shot paradigm for NAS due to its high efficiency. This work turns to zero-order optimization and proposes a novel NAS scheme, called ZARTS, to search without enforcing the above approximation. In particular, results on 12 benchmarks verify the outstanding robustness of ZARTS, where the performance of DARTS collapses due to its known instability issue.
  arXiv  Detail & Related papers  (2021-10-10T09:35:15Z)
• Constructing Stronger and Faster Baselines for Skeleton-based Action Recognition [19.905455701387194]
  We present an efficient Graph Convolutional Network (GCN) baseline for skeleton-based action recognition. On two large-scale datasets, i.e., NTU RGB+D 60 and 120, the proposed EfficientGCN-B4 baseline outperforms other State-Of-The-Art (SOTA) methods.
  arXiv  Detail & Related papers  (2021-06-29T07:09:11Z)
• FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining [65.39532971991778]
  We present an accuracy predictor that scores architecture and training recipes jointly, guiding both sample selection and ranking. We run fast evolutionary searches in just CPU minutes to generate architecture-recipe pairs for a variety of resource constraints. FBNetV3 makes up a family of state-of-the-art compact neural networks that outperform both automatically and manually-designed competitors.
  arXiv  Detail & Related papers  (2020-06-03T05:20:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.