New generative framework creates realistic synthetic populations for travel models

Researchers have developed a novel two-stage generative framework to create realistic synthetic populations for activity-based travel demand models. The first stage uses a regularized Wasserstein GAN with gradient penalty (WGAN-GP) to improve the feasibility, diversity, and novelty of static socio-demographic attributes. The second stage employs Transformer and LSTM-Attention models to generate sequential travel behaviors, such as trip chains, conditioned on the synthesized tabular data. This approach enhances the recovery of valid unseen attribute combinations and improves overall model performance. AI

IMPACT Enhances the realism and utility of synthetic data for transportation modeling, potentially improving urban planning and policy decisions.

RANK_REASON Academic paper detailing a novel generative framework for synthetic population generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• CLAP
• integrated graphics processor
• Low Dynamic Range
• LSTM-Attention
• Transformer
• Wasserstein GAN
• WGAN-GP

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →