Computational Mobility: Learning, Prediction, and Algorithms
summary
Describes production-ready methods for large road graphs: partitioning via random walks plus max flow yields balanced separators that cut query latency and are deployed; electrical-flow decompositions with fast Laplacian solvers generate many alternate routes quickly and robustly; compact distance oracles give rapid approximate travel times; EV routing prototypes handle battery and charger state in-vehicle; online learning with bandit and expert models uses auxiliary probes to reduce routing error and adapt to user preference and traffic feedback.
AI-written summary · about
chapters (6)
- 0:00Start
- 2:25Future Directions in Mobility
- 5:28Universal Transportation Platform
- 8:15Customizable Routing: EV Routing
- 10:31Modeling the Road Network
- 11:36Learning Universal Network Representations
related talks
20mAI on the Fast Lane: Smart Traffic Systems to Unclog our Cities
47mRecent trends in combinatorial optimization augmented machine learning: A graph learning perspective
07mUrban Map Inference by Pervasive Vehicular Sensing Systems with Complementary Mobility
45mKEYNOTE: Traffic simulation modeling: combining generative modeling with transportation science ...
20mRoute Optimization Engine for the Real World
16mOptimized Routing with Satellite Imagery and Computer Vision
28m"Where, When, Why, and How: Directi
08mBuilding Production-Ready ML-Based Network Routing