analiza aiqbench
This research introduces LTV-CTDNet, a machine learning framework designed to forecast traffic turning movements while enforcing physical constraints like non-negative counts. By decomposing predictions into approach totals and proportions, the model avoids the physically impossible outputs often generated by unconstrained neural networks. Although accuracy gains over standard sequence models are modest, the method ensures structural admissibility and interpretability for signal control applications.
LTV-CTDNet: Compositional Turning Decomposition for Short-Term Turning-Movement Forecasting
arXiv:2609.34014v1 Announce Type: new Abstract: Short-term turning-movement forecasts can support signal control and corridor operations, but unconstrained neural networks may produce physically impossible negative counts or outputs that are not explicitly tied to an approach-demand total. This study introduces the Linear Temporal-Variable Compositional Turning Decomposition Network (LTV-CTDNet), a forecasting framework designed to combine competitive accuracy with structurally admissible outputs. LTV-CTDNet was evaluated using seven months of 15-minute LiDAR observations from eight monitored corridor locations in Nashville, Tennessee. Its lightweight encoder combines recent turning-movement history, weekly time-slot embeddings, and location embeddings. The Compositional Turning Decomposition framework separately predicts nonnegative approach totals and within-approach turning proportions, then reconstructs movement forecasts from these components. Among the evaluated predefined configurations, LTV-CTDNet achieved a movement-level MAE of 1.8189 and RMSE of 3.8072. Its accuracy gains over the strongest sequence models were modest, but it produced no negative forecasts, while unconstrained learned models generated negative values in approximately 10.6% to 29.2% of raw forecast cells. The framework enforces nonnegative outputs and exact agreement between each model-predicted approach total and the sum of its component movements by construction, providing directly interpretable forecasts without clipping or coherence correction.
Abstract:Short-term turning-movement forecasts can support signal control and corridor operations, but unconstrained neural networks may produce physically impossible negative counts or outputs that are not explicitly tied to an approach-demand total. This study introduces the Linear Temporal-Variable Compositional Turning Decomposition Network (LTV-CTDNet), a forecasting framework designed to combine competitive accuracy with structurally admissible outputs. LTV-CTDNet was evaluated using seven months of 15-minute LiDAR observations from eight monitored corridor locations in Nashville, Tennessee. Its lightweight encoder combines recent turning-movement history, weekly time-slot embeddings, and location embeddings. The Compositional Turning Decomposition framework separately predicts nonnegative approach totals and within-approach turning proportions, then reconstructs movement forecasts from these components. Among the evaluated predefined configurations, LTV-CTDNet achieved a movement-level MAE of 1.8189 and RMSE of 3.8072. Its accuracy gains over the strongest sequence models were modest, but it produced no negative forecasts, while unconstrained learned models generated negative values in approximately 10.6% to 29.2% of raw forecast cells. The framework enforces nonnegative outputs and exact agreement between each model-predicted approach total and the sum of its component movements by construction, providing directly interpretable forecasts without clipping or coherence correction.
Accepted: September 26, 2026, for presentation at the 2027 TRB Annual Meeting, Washington, D.C. (Paper No. TRBAM-27-06161)
Machine Learning (cs.LG)
arXiv:2609.34014 [cs.LG]
(or arXiv:2609.34014v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.34014 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 296 cuvinte
Subiecte conexe
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · 12 iun. 2026
Does preservation make sense before we know how to revive?
LessWrong · 15 iun. 2026
Koopman Theory and Metaethics
LessWrong · 19 sept. 2026
Off-Target Effects of Response-Style Alignment in a Korean 27B Language Model
arXiv · 12 sept. 2026