Saved in:
Bibliographic Details
Main Authors: Ryu, Seunghee, Kwon, Donghoon, Choi, Seongjin, Deshwal, Aryan, Kang, Seungmo, Osorio, Carolina
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.18824
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909862414254080
author Ryu, Seunghee
Kwon, Donghoon
Choi, Seongjin
Deshwal, Aryan
Kang, Seungmo
Osorio, Carolina
author_facet Ryu, Seunghee
Kwon, Donghoon
Choi, Seongjin
Deshwal, Aryan
Kang, Seungmo
Osorio, Carolina
contents We introduce \textbf{BO4Mob}, a new benchmark framework for high-dimensional Bayesian Optimization (BO), driven by the challenge of origin-destination (OD) travel demand estimation in large urban road networks. Estimating OD travel demand from limited traffic sensor data is a difficult inverse optimization problem, particularly in real-world, large-scale transportation networks. This problem involves optimizing over high-dimensional continuous spaces where each objective evaluation is computationally expensive, stochastic, and non-differentiable. BO4Mob comprises five scenarios based on real-world San Jose, CA road networks, with input dimensions scaling up to 10,100. These scenarios utilize high-resolution, open-source traffic simulations that incorporate realistic nonlinear and stochastic dynamics. We demonstrate the benchmark's utility by evaluating five optimization methods: three state-of-the-art BO algorithms and two non-BO baselines. This benchmark is designed to support both the development of scalable optimization algorithms and their application for the design of data-driven urban mobility models, including high-resolution digital twins of metropolitan road networks. Code and documentation are available at https://github.com/UMN-Choi-Lab/BO4Mob.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18824
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BO4Mob: Bayesian Optimization Benchmarks for High-Dimensional Urban Mobility Problem
Ryu, Seunghee
Kwon, Donghoon
Choi, Seongjin
Deshwal, Aryan
Kang, Seungmo
Osorio, Carolina
Machine Learning
We introduce \textbf{BO4Mob}, a new benchmark framework for high-dimensional Bayesian Optimization (BO), driven by the challenge of origin-destination (OD) travel demand estimation in large urban road networks. Estimating OD travel demand from limited traffic sensor data is a difficult inverse optimization problem, particularly in real-world, large-scale transportation networks. This problem involves optimizing over high-dimensional continuous spaces where each objective evaluation is computationally expensive, stochastic, and non-differentiable. BO4Mob comprises five scenarios based on real-world San Jose, CA road networks, with input dimensions scaling up to 10,100. These scenarios utilize high-resolution, open-source traffic simulations that incorporate realistic nonlinear and stochastic dynamics. We demonstrate the benchmark's utility by evaluating five optimization methods: three state-of-the-art BO algorithms and two non-BO baselines. This benchmark is designed to support both the development of scalable optimization algorithms and their application for the design of data-driven urban mobility models, including high-resolution digital twins of metropolitan road networks. Code and documentation are available at https://github.com/UMN-Choi-Lab/BO4Mob.
title BO4Mob: Bayesian Optimization Benchmarks for High-Dimensional Urban Mobility Problem
topic Machine Learning
url https://arxiv.org/abs/2510.18824