Comparative Evaluation of Neural Network Architectures for Generalizable Human Spatial Preference Prediction in Unseen Built Environments
This work addresses the problem of model generalizability for human preference prediction in Cyber-Physical-Social Infrastructure Systems, though it is incremental as it focuses on a comparative evaluation using synthetic data.
The study compared Graph Neural Networks, Convolutional Neural Networks, and feedforward Neural Networks for predicting human spatial preferences in unseen built environments, finding that Graph Neural Networks achieved the highest generalizability score based on area under the precision-recall curve.
The capacity to predict human spatial preferences within built environments is instrumental for developing Cyber-Physical-Social Infrastructure Systems (CPSIS). A significant challenge in this domain is the generalizability of preference models, particularly their efficacy in predicting preferences within environmental configurations not encountered during training. While deep learning models have shown promise in learning complex spatial and contextual dependencies, it remains unclear which neural network architectures are most effective at generalizing to unseen layouts. To address this, we conduct a comparative study of Graph Neural Networks, Convolutional Neural Networks, and standard feedforward Neural Networks using synthetic data generated from a simplified and synthetic pocket park environment. Beginning with this illustrative case study, allows for controlled analysis of each model's ability to transfer learned preference patterns to unseen spatial scenarios. The models are evaluated based on their capacity to predict preferences influenced by heterogeneous physical, environmental, and social features. Generalizability score is calculated using the area under the precision-recall curve for the seen and unseen layouts. This generalizability score is appropriate for imbalanced data, providing insights into the suitability of each neural network architecture for preference-aware human behavior modeling in unseen built environments.