# Prediction of rare feature combinations in population synthesis: Application of deep generative modelling

arXiv: https://arxiv.org/abs/1909.07689 · HTML: https://mlsanity.com/paper/1909.07689/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Sep 2019.

## Built on by (most cited first, 19 of 19)

- [Choice modelling in the age of machine learning -- discussion paper](https://mlsanity.com/paper/2101.11948/) (2021)
- [A Deep Generative Model for Feasible and Diverse Population Synthesis](https://mlsanity.com/paper/2208.01403/) (2022)
- [Copula-based transferable models for synthetic population generation](https://mlsanity.com/paper/2302.09193/) (2023)
- [A Large Language Model for Feasible and Diverse Population Synthesis](https://mlsanity.com/paper/2505.04196/) (2025)
- [Leveraging Generative AI for Urban Digital Twins: A Scoping Review on the Autonomous Generation of Urban Data, Scenarios, Designs, and 3D City Models for Smart City Advancement](https://mlsanity.com/paper/2405.19464/) (2024)
- [A Gentle Introduction and Tutorial on Deep Generative Models in Transportation Research](https://mlsanity.com/paper/2410.07066/) (2024)
- [Generating Feasible and Diverse Synthetic Populations Using Diffusion Models](https://mlsanity.com/paper/2508.09164/) (2025)
- [Robustness Analysis of Deep Learning Models for Population Synthesis](https://mlsanity.com/paper/2211.13339/) (2022)
- [Generating Spatial Synthetic Populations Using Wasserstein Generative Adversarial Network: A Case Study with EU-SILC Data for Helsinki and Thessaloniki](https://mlsanity.com/paper/2501.16080/) (2025)
- [Exact Synthetic Populations for Scalable Societal and Market Modeling](https://mlsanity.com/paper/2512.07306/) (2025)
- [Population Synthesis using Incomplete Information](https://mlsanity.com/paper/2510.00859/) (2025)
- [Target Population Synthesis using CT-GAN](https://mlsanity.com/paper/2510.00871/) (2025)
- [Population synthesis with geographic coordinates](https://mlsanity.com/paper/2510.09669/) (2025)
- [SemaPop: Semantic-Persona Conditioned and Controllable Population Synthesis](https://mlsanity.com/paper/2602.11569/) (2026)
- [Enhancing Diversity and Feasibility: Joint Population Synthesis from Multi-source Data Using Generative Models](https://mlsanity.com/paper/2602.15270/) (2026)
- [A Bayesian Framework for Evaluating Scenario Compatibility in Generative Population Synthesis](https://mlsanity.com/paper/2607.03190/) (2026)
- [Feasible and Novel Synthetic Population Generation with Tabular and Sequential Travel Attributes](https://mlsanity.com/paper/2608.15867/) (2026)
- [LEBGen: An LLM-Enhanced Bayesian Network Framework for Few-Shot Travel Survey Data Generation](https://mlsanity.com/paper/2609.08288/) (2026)
- [Masked Diffusion for Travel Diaries Synthesis](https://mlsanity.com/paper/2610.17109/) (2026)

## Abstract (arXiv)

In population synthesis applications, when considering populations with many attributes, a fundamental problem is the estimation of rare combinations of feature attributes. Unsurprisingly, it is notably more difficult to reliably representthe sparser regions of such multivariate distributions and in particular combinations of attributes which are absent from the original sample. In the literature this is commonly known as sampling zeros for which no systematic solution has been proposed so far. In this paper, two machine learning algorithms, from the family of deep generative models,are proposed for the problem of population synthesis and with particular attention to the problem of sampling zeros. Specifically, we introduce the Wasserstein Generative Adversarial Network (WGAN) and the Variational Autoencoder(VAE), and adapt these algorithms for a large-scale population synthesis application. The models are implemented on a Danish travel survey with a feature-space of more than 60 variables. The models are validated in a cross-validation scheme and a set of new metrics for the evaluation of the sampling-zero problem is proposed. Results show how these models are able to recover sampling zeros while keeping the estimation of truly impossible combinations, the structural zeros, at a comparatively low level. Particularly, for a low dimensional experiment, the VAE, the marginal sampler and the fully random sampler generate 5%, 21% and 26%, respectively, more structural zeros per sampling zero generated by the WGAN, while for a high dimensional case, these figures escalate to 44%, 2217% and 170440%, respectively. This research directly supports the development of agent-based systems and in particular cases where detailed socio-economic or geographical representations are required.