AlphaTransit pairs MCTS with a learned policy-value network to reach 54.6% and 82.1% service rates on a Bloomington transit benchmark, outperforming plain RL and plain MCTS baselines.
arXiv preprint arXiv:2405.10074 , year=
3 Pith papers cite this work, alongside 2 external citations. Polarity classification is still indexing.
3
Pith papers citing it
2
external citations · external index
years
2026 3verdicts
UNVERDICTED 3representative citing papers
An iterative exact algorithm solves a mixed-integer line planning model faster than CPLEX by dynamically expanding paths and frequencies, and accounting for lost demand improves overall resource efficiency.
Empirical comparison on Dutch and Swiss railway instances shows a compact direct connection model outperforms the canonical change-and-go network on over 83% of 972 cases, with the latter failing on many large instances.
citing papers explorer
-
AlphaTransit: Learning to Design City-scale Transit Routes
AlphaTransit pairs MCTS with a learned policy-value network to reach 54.6% and 82.1% service rates on a Bloomington transit benchmark, outperforming plain RL and plain MCTS baselines.
-
An Exact Algorithm for Public Transport Line Planning Considering Passenger and Operational Costs and Lost Demand
An iterative exact algorithm solves a mixed-integer line planning model faster than CPLEX by dynamically expanding paths and frequencies, and accounting for lost demand improves overall resource efficiency.
-
Line Planning at Scale: Models, Methods, and Insights
Empirical comparison on Dutch and Swiss railway instances shows a compact direct connection model outperforms the canonical change-and-go network on over 83% of 972 cases, with the latter failing on many large instances.