An Information-Theoretic Approach for Estimating Scenario Generalization in Crowd Motion Prediction
This work addresses the challenge of efficient model evaluation for researchers and practitioners in crowd motion prediction, though it is incremental as it builds on existing generalization estimation methods.
The paper tackles the problem of time-consuming model selection and parameter tuning in crowd motion prediction by proposing a training-free, model-agnostic scoring method called ISDQ to estimate scenario generalization, which is validated on simulated and real-world tasks for selecting optimal domain pairs before training.
Learning-based approaches to modeling crowd motion have become increasingly successful but require training and evaluation on large datasets, coupled with complex model selection and parameter tuning. To circumvent this tremendously time-consuming process, we propose a novel scoring method, which characterizes generalization of models trained on source crowd scenarios and applied to target crowd scenarios using a training-free, model-agnostic Interaction + Diversity Quantification score, ISDQ. The Interaction component aims to characterize the difficulty of scenario domains, while the diversity of a scenario domain is captured in the Diversity score. Both scores can be computed in a computation tractable manner. Our experimental results validate the efficacy of the proposed method on several simulated and real-world (source,target) generalization tasks, demonstrating its potential to select optimal domain pairs before training and testing a model.