Adaptive bidirectional spatial-temporal prediction model for traffic speed in large-scale road networks
Bibliographic record
Abstract
Large-scale road network traffic speed prediction plays a critical role in urban computing tasks and ensures the smooth flow of city traffic. Graph Convolutional Networks (GCNs) have natural advantages in representing non-Euclidean data. However, Laplacian-based GCNs are built on the assumption of an undirected graph, which is inconsistent with the directed graph formed by large-scale road traffic networks represented by sensors. To this end, we propose a novel Adaptive Bidirectional Spatial-Temporal Network (ABSTN) for urban traffic speed prediction. Specifically, we develop an Adaptive Bidirectional Graph Convolutional Unit (ABGC). On the one hand, ABGC maintains 2 heterogeneous embedding dictionaries to learn the potential pairwise relationships between nodes. On the other hand, ABGC simultaneously performs GCN operations on out-/in-degree to capture the up-/down-stream relationships of traffic flow. Subsequently, ABGC acts as a linear layer is embedded in Gated Recurrent Units (GRUs) to jointly capture spatial-temporal dependencies. Furthermore, we introduce an Interactive Multi-head Attention block (IMA) within the encoder-decoder framework to achieve long-range dependency modeling in the temporal dimension. Finally, a scheduled sampling scheme is employed to enhance the model’s generalization for multi-step prediction. Extensive experiments on two real-world traffic speed datasets demonstrate that the proposed ABSTN achieves state-of-the-art performance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".