A Gentle Introduction and Tutorial on Deep Generative Models in Transportation Research
arXiv:2410.07066 · doi:10.1016/j.trc.2025.105145
Abstract
Deep Generative Models (DGMs) have rapidly advanced in recent years, becoming essential tools in various fields due to their ability to learn complex data distributions and generate synthetic data. Their importance in transportation research is increasingly recognized, particularly for applications like traffic data generation, prediction, and feature extraction. This paper offers a comprehensive introduction and tutorial on DGMs, with a focus on their applications in transportation. It begins with an overview of generative models, followed by detailed explanations of fundamental models, a systematic review of the literature, and practical tutorial code to aid implementation. The paper also discusses current challenges and opportunities, highlighting how these models can be effectively utilized and further developed in transportation research. This paper serves as a valuable reference, guiding researchers and practitioners from foundational knowledge to advanced applications of DGMs in transportation research.
64 pages, 21 figures, 4 tables
References in corpus (38)
- Auto-Encoding Variational Bayes
- Conditional Generative Adversarial Nets
- Denoising Diffusion Probabilistic Models
- NICE: Non-linear Independent Components Estimation
- Score-Based Generative Modeling through Stochastic Differential Equations
- Generative Modeling by Estimating Gradients of the Data Distribution
- The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
- External Validity: From Do-Calculus to Transportability Across Populations
- Tackling the Generative Learning Trilemma with Denoising Diffusion GANs
- Modeling Human Driving Behavior through Generative Adversarial Imitation Learning
- Scalable Population Synthesis with Deep Generative Modeling
- Probabilistic Crowd GAN: Multimodal Pedestrian Trajectory Prediction using a Graph Vehicle-Pedestrian Attention Network
- A Generative Car-following Model Conditioned On Driving Styles
- Multivariate Probabilistic Time Series Forecasting via Conditioned Normalizing Flows
- D-GAN: Deep Generative Adversarial Nets for Spatio-Temporal Prediction
- Generative machine learning methods for multivariate ensemble post-processing
- Collapse by Conditioning: Training Class-conditional GANs with Limited Data
- DiffTraj: Generating GPS Trajectory with Diffusion Probabilistic Model
- DriveSceneGen: Generating Diverse and Realistic Driving Scenarios from Scratch
- SurrealDriver: Designing LLM-powered Generative Driver Agent Framework based on Human Drivers' Driving-thinking Data
- Probabilistic Forecasting with Generative Networks via Scoring Rule Minimization
- A Survey of Generative AI for Intelligent Transportation Systems: Road Transportation Perspective
- Diffusion-Based Environment-Aware Trajectory Prediction
- Deep Activity Model: A Generative Approach for Human Mobility Pattern Synthesis
- SceneDiffuser: Efficient and Controllable Driving Simulation Initialization and Rollout
- CVAE-H: Conditionalizing Variational Autoencoders via Hypernetworks and Trajectory Forecasting for Autonomous Driving
- SpecSTG: A Fast Spectral Diffusion Framework for Probabilistic Spatio-Temporal Traffic Forecasting
- DPTraj-PM: Differentially Private Trajectory Synthesis Using Prefix Tree and Markov Process
- Origin-Destination Network Generation via Gravity-Guided GAN
- Transfusor: Transformer Diffusor for Controllable Human-like Generation of Vehicle Lane Changing Trajectories
- Crash Data Augmentation Using Conditional Generative Adversarial Networks (CGAN) for Improving Safety Performance Functions
- WcDT: World-centric Diffusion Transformer for Traffic Scene Generation
- Versatile Behavior Diffusion for Generalized Traffic Agent Simulation
- Distributional Diffusion Models with Scoring Rules
- Exploring Attention GAN for Vehicle Motion Prediction
- A Diffusion-Model of Joint Interactive Navigation
- Complexity-aware Large Scale Origin-Destination Network Generation via Diffusion Model
- Robustness Analysis of Deep Learning Models for Population Synthesis