A bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data

Abstract

The emergence of data-driven demand analysis have led to the increased use of generative modelling to learn the probabilistic dependencies between random variables. Although their apparent use has largely been limited to image recognition and classification in recent years, generative machine learning algorithms can be a powerful tool for travel behaviour research by replicating travel behaviour by the underlying properties of data structures. In this paper, we examine the use of generative machine learning approach for analyzing multiple discrete-continuous (MDC) travel behaviour data. We provide a plausible perspective of how we can exploit the use of machine learning techniques to interpret the underlying heterogeneities in the data. We show that generative models are conceptually similar to choice selection behaviour process through information entropy and variational Bayesian inference. Without loss of generality, we consider a restricted Boltzmann machine (RBM) based algorithm with multiple discrete-continuous layer, formulated as a variational Bayesian inference optimization problem. We systematically describe the proposed machine learning algorithm and develop a process of analyzing travel behaviour data from a generative learning perspective. We show parameter stability from model analysis and simulation tests on an open dataset with multiple discrete-continuous dimensions from a data size of 293,330 observations. For interpretability, we derive the conditional probabilities, elasticities and perform statistical analysis on the latent variables. We show that our model can generate statistically similar data distributions for travel forecasting and prediction and performs better than purely discriminative methods in validation. Our results indicate that latent constructs in generative models can accurately represent the joint distribution consistently on MDC data.

Decoding pedestrian and automated vehicle interactions using immersive virtual reality and interpretable deep learning

“…However, disaggregate information for individuals still lacks the reliability acquired by discrete choice models. Wong and Farooq (2020) used Jacobian determinant of generative models to calculate the elasticity of mode choice with respect to different explanatory variables. Jacobian matrix, in this study, was generated for each instance of all the conditional outputs, and density of elasticises was estimated across the data points.…”

Choice modelling in the age of machine learning - Discussion paper

“…If we can make this progress, we believe the future of our field would be brighter than ever. (Xie et al 2003;Zhang and Xie 2008;Tortum et al 2009;Lu and Kawamura 2010;Omrani et al 2013;Hagenauer and Helbich 2017;Wong et al 2017;Alwosheel et al 2018;Lee et al 2018;Shi and Yin 2018;Sun et al 2018;Wang and Ross 2018;Alwosheel et al 2019;Lee et al 2019;Lhéritier et al 2019;Paz et al 2019;Van Cranenburgh and Alwosheel 2019;Zhao et al 2019;Newman and Garrow 2020;Wang et al 2020a;Wang et al 2020b;Wong and Farooq 2020;Yao and Bekhor 2020;Zhu et al 2021)…”

Travel Behavior Modeling: Taxonomy, Challenges, and Opportunities

“…3) Restricted Boltzmann Machine: This algorithm enables consideration of information heterogeneity and variable correlations in processing that commonly occurs on large scale datasets. A recent study investigated on generative machine learning approaches by analyzing multiple discrete-continuous (MDC) travel behavior event from 293,330 travel data [19]. RBM-based algorithm performs significantly better in modeling behavior with larger datasets due to increased modeling accuracy and future forecasting.…”