Skip to content
Preprint

RiskTraf: Risk-Extrapolated Residual Learning for Multi-Variate Traffic Flow Prediction

Aug 2026 · 0 citations · 51 references
Computer Science

TL;DR

PEMSB-3V, a public benchmark suite that preserves raw flow, speed, and occupancy measurements from PeMS detectors for flow prediction, and RiskTraf, a model-agnostic risk-extrapolated residual plug-in that consistently improves diverse forecasting backbones and outperforms debiasing and distribution-shift adaptation methods.

Abstract

Traffic sensors commonly record flow, speed, and occupancy, but standard traffic flow forecasting benchmarks and models rarely exploit all three raw measurements reliably. Although speed and occupancy provide sensor-native traffic-state information beyond flow alone, existing releases often omit these variables, replace them with proxies, or contain logically inconsistent records. Moreover, direct empirical risk minimization over three-variable inputs may exploit regime-dependent shortcuts, as the relationships among flow, speed, and occupancy vary substantially between free-flow and congested states. We introduce \textbf{PEMSB-3V}, a public benchmark suite that preserves raw flow, speed, and occupancy measurements from PeMS detectors for flow prediction. We also propose \textbf{RiskTraf}, a model-agnostic risk-extrapolated residual plug-in. For each trained spatio-temporal backbone, RiskTraf freezes the selected checkpoint and learns a lightweight zero-start residual head from historical speed and occupancy. The residual head constructs ordered traffic-risk environments and optimizes horizon-wise flow corrections with a risk extrapolation objective, thereby mitigating regime-specific shortcut correlations without modifying the backbone. Extensive experiments demonstrate that RiskTraf consistently improves diverse forecasting backbones and outperforms debiasing and distribution-shift adaptation methods. Our code and benchmark are available at https://github.com/Guangyu4/RiskTraf.

View source

Similar papers

Open access Sep 2026

MURECAST: Memory-Based Utility-Aligned Residual Evidence with Calibrated Activation for Selective Correction of Traffic-Flow Forecasts

This work proposes Memory-Based Utility-Aligned Residual Evidence with Calibrated Activation for Selective Correction of Traffic-Flow Forecasts (MURECAST), which treats historical residuals as candidate interventions and ranks first in dataset–metric comparisons.

Xue-Ting Jin, Jin-Feng Xu, Qian-Xin Xie et al. · 0 citations

HALT: Human-in-the-Loop Data Acquisition for Supervised Traffic Causality Discovery

HALT, a human-in-the-loop label acquisition framework for event-based traffic causality discovery, is presented and it is suggested that metric-aware active labeling can make supervised traffic-causality discovery practical under limited expert-labeling budgets.

Unknown authors · 0 citations
#machine learning Preprint Sep 2026

FlowATC: Aircraft Trajectory Prediction via Flow Matching

A flow-matching architecture trained exclusively on historical aircraft trajectories, with no route labels or chart supervision, that generates aircraft trajectory distributions that closely match historical traffic, reproducing known airspace structure around San Francisco Airport.

Mathurin Petit, Emir Torun, Louis Brusset et al. · 0 citations
Dec 2026

Short-Term Traffic Speed Prediction Integrating Traffic Flow Theory for Enhanced Urban Congestion Mitigation

A theory guided deep learning framework that integrates traffic flow theory with short-term speed prediction and allows operators to act early with measures such as variable speed limits before congestion fully develops, shifting traffic management from reactive monitoring toward proactive control.

Donghyeok Park, Yoon-Young Choi, Juneyoung Park · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.