analiza aiqbench
This research introduces RUMnets, a neural network architecture designed to rigorously approximate Random Utility Maximization models while maintaining theoretical consistency with the underlying economic principles of discrete choice. By deriving bounds on generalization error and validating performance against standard machine learning methods, the study bridges deep learning capabilities with formal econometric frameworks for predicting agent behavior.
Representing Random Utility Choice Models with Neural Networks
arXiv:2207.12877v3 Announce Type: replace Abstract: Motivated by the successes of deep learning, we propose a class of neural network-based discrete choice models, called RUMnets, inspired by the random utility maximization (RUM) framework. This model formulates the agents' random utility function using a sample average approximation. We show that RUMnets sharply approximate the class of RUM discrete choice models: any model derived from random utility maximization has choice probabilities that can be approximated arbitrarily closely by a RUMnet. Reciprocally, any RUMnet is consistent with the RUM principle. Our approach is closely related to ranking-based models and mixtures of multinomial logits proposed in previous literature, in a more general contextual setting. We derive an upper bound on the generalization error of RUMnets fitted on choice data, and provide theoretical insights on their ability to predict choices on new, unseen data depending on critical parameters of the dataset and architecture. The models are estimated by leveraging open-source libraries for training neural networks. We find that RUMnets are competitive against several choice modeling and machine learning methods in terms of predictive accuracy on two real-world datasets. We also conduct synthetic experiments that isolate the effects of each component of the architecture.
Abstract:Motivated by the successes of deep learning, we propose a class of neural network-based discrete choice models, called RUMnets, inspired by the random utility maximization (RUM) framework. This model formulates the agents' random utility function using a sample average approximation. We show that RUMnets sharply approximate the class of RUM discrete choice models: any model derived from random utility maximization has choice probabilities that can be approximated arbitrarily closely by a RUMnet. Reciprocally, any RUMnet is consistent with the RUM principle. Our approach is closely related to ranking-based models and mixtures of multinomial logits proposed in previous literature, in a more general contextual setting. We derive an upper bound on the generalization error of RUMnets fitted on choice data, and provide theoretical insights on their ability to predict choices on new, unseen data depending on critical parameters of the dataset and architecture. The models are estimated by leveraging open-source libraries for training neural networks. We find that RUMnets are competitive against several choice modeling and machine learning methods in terms of predictive accuracy on two real-world datasets. We also conduct synthetic experiments that isolate the effects of each component of the architecture.
Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
arXiv:2207.12877 [cs.LG]
(or arXiv:2207.12877v3 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2207.12877 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 291 cuvinte
