Combining Discrete Choice Models and Neural Networks through Embeddings: Formulation, Interpretability and Performance (2021-09-24T00:00:00.000000Z)

TL;DR

This study uses continuous vector representations, called embeddings, for encoding categorical or discrete explanatory variables with a special focus on interpretability and model transparency, and delivers state-of-the-art predictive performance, outperforming existing ANN-based models while drastically reducing the number of required network parameters.

Authors

Francisco Câmara Pereira

2 papers

Ioanna Arkoudi

1 papers

C. L. Azevedo

1 papers

TL;DR

Authors

References53 items

Why did you predict that? Towards explainable artificial neural networks for travel demand analysis

ResLogit: A residual neural network logit model for data-driven choice modelling

Enhancing discrete choice models with representation learning

Multi-Task Learning with Deep Neural Networks: A Survey

Deep neural networks for choice analysis: Architecture design with alternative-specific utility functions

A neural-embedded discrete choice model: Learning taste representation with strengthened interpretability

Urban2Vec: Incorporating Street View Imagery and POIs for Multi-Modal Urban Neighborhood Embedding

Revealing the hidden features in traffic prediction via entity embedding

Rethinking travel behavior modeling representations through embeddings

Opening Up the Conversation: Topic Modeling for Automated Text Analysis in Travel Surveys

Evaluation of sentence embeddings in downstream and linguistic probing tasks

From Word to Sense Embeddings: A Survey on Vector Representations of Meaning

From ITDL to Place2Vec: Reasoning About Place Type Similarity and Relatedness by Learning Embeddings From Augmented Spatial Contexts

A Unified Approach to Interpreting Model Predictions

Machine Learning: An Applied Econometric Approach

POI2Vec: Geographical Latent Representation for Predicting Future Visitors

node2vec: Scalable Feature Learning for Networks

Asymmetric, closed-form, finite-parameter models of multinomial choice

GT-SEER: Geo-Temporal SEquential Embedding Rank for Point-of-interest Recommendation

From Twitter to detector: real-time traffic incident detection using social media data

Entity Embeddings of Categorical Variables

“Why Should I Trust You?”: Explaining the Predictions of Any Classifier

Deep Residual Learning for Image Recognition

Expanding n-gram training data for language models based on morpho-syntactic transformations

Artificial Neural Networks Applied to Taxi Destination Prediction

Adam: A Method for Stochastic Optimization

Streetscore -- Predicting the Perceived Safety of One Million Streetscapes

DeepWalk: online learning of social representations

Using semi-open questions to integrate perceptions in choice models

DeViSE: A Deep Visual-Semantic Embedding Model

Translating Embeddings for Modeling Multi-relational Data

The Collaborative Image of The City: Mapping the Inequality of Urban Perception

Sample size requirements for stated choice experiments

Efficient Estimation of Word Representations in Vector Space

A Monte Carlo experiment to analyze the curse of dimensionality in estimating random coefficients models with a full variance-covariance matrix

A Three-Way Model for Collective Learning on Multi-Relational Data

Dynamic Discrete Choice Models for Transportation

Word Representations: A Simple and General Method for Semi-Supervised Learning

Transfer Learning via Dimensionality Reduction

An empirical comparison of the validity of a neural net based multinomial logit choice model to alternative model specifications

A flexible brand choice model based on neural net methodology A comparison to the linear utility multinomial logit model and its latent class extension

Hybrid Choice Models: Progress and Challenges

Neural networks and the multinomial logit for brand choice modelling: a hybrid approach

Analysis of injury severity and vehicle occupancy in truck- and non-truck-involved accidents.

An exploratory multinomial logit analysis of single-vehicle motorcycle accident severity

Regularized Categorical Embedding for Effective Demand Forecasting of Bike Sharing System

Explaining subjective perceptions of public spaces as a function of the built environment: A massive data approach

Analysis of Household Survey Sample Size in Trip Modelling Process

RAND Corporation

High Dimensional Nonparametric Discrete Choice Model

Multi-task feature learning

Acceptance of modal innovation: the case of the SwissMetro

On the mean accuracy of statistical pattern recognizers

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names