arXiv subjects

Francisco Camara Pereira

Publications and source records attributed to Francisco Camara Pereira.

18 recordsLinked to original sources

arXiv2026

A Response Theory Probe for Learned Stochastic AI Simulators, Tested on Lorenz-63

Machine-learning emulators of chaotic and stochastic systems are usually validated on forecast skill and long-run statistics. Neither certifies that an emulator responds correctly to forcing, the property that projection and attribution studies rely on. Linear response theory makes this testable: the forced response follows from unperturbed correlations through a generalized fluctuation-dissipation relation, and decomposes over the stochastic Ruelle-Pollicott resonances of the Koopman generator. Building on the Koopmanism Response framework, we turn this into a calibrated, mode-resolved test for learned surrogates: each surrogate rollout passes or fails each check, and failure rates are compared with those of independent realizations of the true system. On stochastic Lorenz-63, a three-variable toy model, we evaluate SINDy, an MLP, a reservoir computer, a neural ODE and a neural SDE with learned diffusion, over up to 80 rollouts each. A sparse-regression model with the correct library passes every check at rates consistent with the true system. Invariant-statistics fidelity and response fidelity dissociate in both directions: a quarter of reservoir-computer rollouts pass every invariant-statistics check and match the static susceptibility $χ(0)$, yet misrepresent the slow relaxation modes, while the neural ODE and SDE rarely meet the invariant-statistics floor but recover those modes in three quarters of rollouts. As expected of a time-integrated quantity dominated here by fast relaxation, $χ(0)$ does not separate these cases. For a fixed network, the training formulation (one-step drift, flow map, or multi-step through the integrator) decides which of these properties it gets right.

math.DS↗

arXiv2026

Climate Surrogates for Scalable Multi-Agent Reinforcement Learning: A Case Study with CICERO-SCM

Climate policy analysis requires models that capture multi-gas climate effects, but such models are too slow to embed in reinforcement learning loops at scale. In collaboration with the European Environment Agency, we develop a multi-agent reinforcement learning (MARL) framework that integrates a higher-fidelity climate surrogate as the environment transition, enabling regional agents to learn policies under multi-gas dynamics. We train a recurrent surrogate on $20{,}000$ multi-gas emission pathways to emulate CICERO-SCM. The surrogate achieves near-simulator accuracy (global-mean temperature RMSE $\approx\!4\!\times \!10^{-4}\,\mathrm{K}$) with $\sim\!1000\times$ faster one-step inference and yields $>\!100\times$ end-to-end MARL training speed-up. We show policy agreement with the simulator in tractable settings and propose a replay- and rank-consistency test (Kendall's $τ$) for assessing policy fidelity when simulator-in-the-loop training is infeasible. This enables large-scale multi-agent policy experiments while retaining high-fidelity multi-gas climate response.

cs.LG↗

arXiv2026

Learning to Learn the Macroscopic Fundamental Diagram using Physics-Informed and meta Machine Learning techniques

The Macroscopic Fundamental Diagram is a popular tool used to describe traffic dynamics in an aggregated way, with applications ranging from traffic control to incident analysis. However, estimating the MFD for a given network requires large numbers of loop detectors, which is not always available in practise. This article proposes a framework to alleviate the data scarcity challenge harnessing Meta-Learning, a subcategory of Machine Learning that trains models to understand and adapt to new tasks on their own. We use Meta-Learning to identify and exploit transferable patterns from data-rich cities to cities where not enough data is available to estimate the MFD. The developed model is trained and tested by leveraging data from multiple cities and exploiting it to model the MFD of other cities with different shares of detectors and topological structures. The proposed Meta-Learning framework is applied to an ad-hoc Multi-Task Physics-Informed Neural Network, specifically designed to estimate the MFD. Results show an average MAE improvement in flow prediction of around 50% across cities (depending on the subset of loop detectors tested). The Meta-Learning framework thus successfully generalises across diverse urban settings and improves performance on cities with limited data, demonstrating the potential of using Meta-Learning when a limited number of detectors is available. We directly test this assumption by applying the Meta-Learning outputs to unseen cities to simulate a real-life application scenario and the wide applicability of the proposed methodology. Finally, the proposed framework is validated against traditional Transfer Learning approaches and tested with FitFun, a model for FD estimation from the literature, to prove its transferability.

cs.LG↗

arXiv2026

Metalearning traffic assignment for network disruptions with graph convolutional neural networks

Building machine-learning models for estimating traffic flows from OD matrices requires an appropriate design of the training process and a training dataset spanning over multiple regimes and dynamics. As machine-learning models rely heavily on historical data, their predictions are typically accurate only when future traffic patterns resemble those observed during training. However, their performance often degrades when there is a significant statistical discrepancy between historical and future conditions. This issue is particularly relevant in traffic forecasting when predictions are required for modified versions of the network, where the underlying graph structure changes due to events such as maintenance, public demonstrations, flooding, or other extreme disruptions. Ironically, these are precisely the situations in which reliable traffic predictions are most needed. In the presented work, we combine a machine-learning model (graph convolutional neural network) with a meta-learning architecture to train the former to quickly adapt to new graph structures and demand patterns, so that it may easily be applied to scenarios in which changes in the road network (the graph) and the demand (the node features) happen simultaneously. Our results show that the use of meta-learning allows the graph neural network to quickly adapt to unseen graphs (network closures) and OD matrixes while easing the burden of designing a training dataset that covers all relevant patterns for the practitioners. The proposed architecture achieves a R^2 of around 0.85 over unseen closures and OD matrixes.

cs.LG↗

arXiv2026

A Foundational Individual Mobility Prediction Model based on Open-Source Large Language Models

Individual mobility prediction plays a key role in urban transport, enabling personalized service recommendations and effective travel management. It is widely modeled by data-driven methods such as machine learning, deep learning, as well as classical econometric methods to capture key features of mobility patterns. However, such methods are hindered in promoting further transferability and robustness due to limited capacity to learn mobility patterns from different data sources, predict in out-of-distribution settings (a.k.a ``zero-shot"). To address this challenge, this paper introduces MoBLLM, a foundational model for individual mobility prediction that aims to learn a shared and transferable representation of mobility behavior across heterogeneous data sources. Based on a lightweight open-source large language model (LLM), MoBLLM employs Parameter-Efficient Fine-Tuning (PEFT) techniques to create a cost-effective training pipeline, avoiding the need for large-scale GPU clusters while maintaining strong performance. We conduct extensive experiments on six real-world mobility datasets to evaluate its accuracy, robustness, and transferability across varying temporal scales (years), spatial contexts (cities), and situational conditions (e.g., disruptions and interventions). MoBLLM achieves the best F1 score and accuracy across all datasets compared with state-of-the-art deep learning models and shows better transferability and cost efficiency than commercial LLMs. Further experiments reveal its robustness under network changes, policy interventions, special events, and incidents. These results indicate that MoBLLM provides a generalizable modeling foundation for individual mobility behavior, enabling more reliable and adaptive personalized information services for transportation management.

cs.CL↗

arXiv2025

Domain-Adapted Pre-trained Language Models for Implicit Information Extraction in Crash Narratives

Free-text crash narratives recorded in real-world crash databases have been shown to play a significant role in improving traffic safety. However, large-scale analyses remain difficult to implement as there are no documented tools that can batch process the unstructured, non standardized text content written by various authors with diverse experience and attention to detail. In recent years, Transformer-based pre-trained language models (PLMs), such as Bidirectional Encoder Representations from Transformers (BERT) and large language models (LLMs), have demonstrated strong capabilities across various natural language processing tasks. These models can extract explicit facts from crash narratives, but their performance declines on inference-heavy tasks in, for example, Crash Type identification, which can involve nearly 100 categories. Moreover, relying on closed LLMs through external APIs raises privacy concerns for sensitive crash data. Additionally, these black-box tools often underperform due to limited domain knowledge. Motivated by these challenges, we study whether compact open-source PLMs can support reasoning-intensive extraction from crash narratives. We target two challenging objectives: 1) identifying the Manner of Collision for a crash, and 2) Crash Type for each vehicle involved in the crash event from real-world crash narratives. To bridge domain gaps, we apply fine-tuning techniques to inject task-specific knowledge to LLMs with Low-Rank Adaption (LoRA) and BERT. Experiments on the authoritative real-world dataset Crash Investigation Sampling System (CISS) demonstrate that our fine-tuned compact models outperform strong closed LLMs, such as GPT-4o, while requiring only minimal training resources. Further analysis reveals that the fine-tuned PLMs can capture richer narrative details and even correct some mislabeled annotations in the dataset.

cs.CL↗

arXiv2025

Learning traffic flows: Graph Neural Networks for Metamodelling Traffic Assignment

The Traffic Assignment Problem is a fundamental, yet computationally expensive, task in transportation modeling, especially for large-scale networks. Traditional methods require iterative simulations to reach equilibrium, making real-time or large-scale scenario analysis challenging. In this paper, we propose a learning-based approach using Message-Passing Neural Networks as a metamodel to approximate the equilibrium flow of the Stochastic User Equilibrium assignment. Our model is designed to mimic the algorithmic structure used in conventional traffic simulators allowing it to better capture the underlying process rather than just the data. We benchmark it against other conventional deep learning techniques and evaluate the model's robustness by testing its ability to predict traffic flows on input data outside the domain on which it was trained. This approach offers a promising solution for accelerating out-of-distribution scenario assessments, reducing computational costs in large-scale transportation planning, and enabling real-time decision-making.

cs.LG↗

arXiv2025

Scenario Discovery for Urban Planning: The Case of Green Urbanism and the Impact on Stress

Urban environments significantly influence mental health outcomes, yet the role of an effective framework for decision-making under deep uncertainty (DMDU) for optimizing urban policies for stress reduction remains underexplored. While existing research has demonstrated the effects of urban design on mental health, there is a lack of systematic scenario-based analysis to guide urban planning decisions. This study addresses this gap by applying Scenario Discovery (SD) in urban planning to evaluate the effectiveness of urban vegetation interventions in stress reduction across different urban environments using a predictive model based on emotional responses collected from a neuroscience-based outdoor experiment in Lisbon. Combining these insights with detailed urban data from Copenhagen, we identify key intervention thresholds where vegetation-based solutions succeed or fail in mitigating stress responses. Our findings reveal that while increased vegetation generally correlates with lower stress levels, high-density urban environments, crowding, and individual psychological traits (e.g., extraversion) can reduce its effectiveness. This work showcases our Scenario Discovery framework as a systematic approach for identifying robust policy pathways in urban planning, opening the door for its exploration in other urban decision-making contexts where uncertainty and design resiliency are critical.

cs.CY↗

arXiv2024

Analyzing the Reporting Error of Public Transport Trips in the Danish National Travel Survey Using Smart Card Data

Household travel surveys have been used for decades to collect individuals and households' travel behavior. However, self-reported surveys are subject to recall bias, as respondents might struggle to recall and report their activities accurately. This study examines the time reporting error of public transit users in a nationwide household travel survey by matching, at the individual level, five consecutive years of data from two sources, namely the Danish National Travel Survey (TU) and the Danish Smart Card system (Rejsekort). Survey respondents are matched with travel cards from the Rejsekort data solely based on the respondents' declared spatiotemporal travel behavior. Approximately, 70% of the respondents were successfully matched with Rejsekort travel cards. The findings reveal a median time reporting error of 11.34 minutes, with an Interquartile Range of 28.14 minutes. Furthermore, a statistical analysis was performed to explore the relationships between the survey respondents' reporting error and their socio-economic and demographic characteristics. The results indicate that females and respondents with a fixed schedule are in general more accurate than males and respondents with a flexible schedule in reporting their times of travel. Moreover, trips reported during weekdays or via the internet displayed higher accuracies compared to trips reported during weekends and holidays or via telephone interviews. This disaggregated analysis provides valuable insights that could help in improving the design and analysis of travel surveys, as well accounting for reporting errors/biases in travel survey-based applications. Furthermore, it offers valuable insights underlying the psychology of travel recall by survey respondents.

stat.AP↗

arXiv2023

Learning to Control Autonomous Fleets from Observation via Offline Reinforcement Learning

Autonomous Mobility-on-Demand (AMoD) systems are an evolving mode of transportation in which a centrally coordinated fleet of self-driving vehicles dynamically serves travel requests. The control of these systems is typically formulated as a large network optimization problem, and reinforcement learning (RL) has recently emerged as a promising approach to solve the open challenges in this space. Recent centralized RL approaches focus on learning from online data, ignoring the per-sample-cost of interactions within real-world transportation systems. To address these limitations, we propose to formalize the control of AMoD systems through the lens of offline reinforcement learning and learn effective control strategies using solely offline data, which is readily available to current mobility operators. We further investigate design decisions and provide empirical evidence based on data from real-world mobility systems showing how offline learning allows to recover AMoD control policies that (i) exhibit performance on par with online methods, (ii) allow for sample-efficient online fine-tuning and (iii) eliminate the need for complex simulation environments. Crucially, this paper demonstrates that offline RL is a promising paradigm for the application of RL-based solutions within economically-critical systems, such as mobility systems.

eess.SY↗

arXiv2023

Determining Causality in Travel Mode Choice

This article presents one of the pioneering studies on causal modeling in travel mode choice decision-making using causal discovery algorithms. These models are a major advancement from conventional correlation-based techniques. We propose a novel methodology that combines causal discovery with structural equation modeling (SEM). This modeling approach overcomes some of the limitations of SEM by combining the strengths of both causal discovery and SEM. Causal discovery algorithms determine causal graphs from observational data and domain knowledge, and SEMs estimate direct causal effects and test the performance of causal discovery algorithms. In this study, we test four causal discovery algorithms: Peter-Clark (PC), Fast Causal Inference (FCI), Fast Greedy Equivalence Search (FGES), and Direct Linear Non-Gaussian Acyclic Models (DirectLiNGAM). The results show that DirectLiNGAM based SEM model best captures causality in mode choice behavior. It passes several goodness-of-fit tests, including Root Mean Square Error of Approximation (RMSEA) and Goodness-of-Fit Index (GFI), and it achieves the lowest Bayesian Information Criterion (BIC) value. The analyses are conducted on data collected from the 2017 National Household Travel Survey in the New York Metropolitan area.

stat.AP↗

arXiv2023

Attitudes and Latent Class Choice Models using Machine learning

Latent Class Choice Models (LCCM) are extensions of discrete choice models (DCMs) that capture unobserved heterogeneity in the choice process by segmenting the population based on the assumption of preference similarities. We present a method of efficiently incorporating attitudinal indicators in the specification of LCCM, by introducing Artificial Neural Networks (ANN) to formulate latent variables constructs. This formulation overcomes structural equations in its capability of exploring the relationship between the attitudinal indicators and the decision choice, given the Machine Learning (ML) flexibility and power in capturing unobserved and complex behavioural features, such as attitudes and beliefs. All of this while still maintaining the consistency of the theoretical assumptions presented in the Generalized Random Utility model and the interpretability of the estimated parameters. We test our proposed framework for estimating a Car-Sharing (CS) service subscription choice with stated preference data from Copenhagen, Denmark. The results show that our proposed approach provides a complete and realistic segmentation, which helps design better policies.

econ.EM↗

arXiv2022

A Neural-embedded Choice Model: TasteNet-MNL Modeling Taste Heterogeneity with Flexibility and Interpretability

Discrete choice models (DCMs) require a priori knowledge of the utility functions, especially how tastes vary across individuals. Utility misspecification may lead to biased estimates, inaccurate interpretations and limited predictability. In this paper, we utilize a neural network to learn taste representation. Our formulation consists of two modules: a neural network (TasteNet) that learns taste parameters (e.g., time coefficient) as flexible functions of individual characteristics; and a multinomial logit (MNL) model with utility functions defined with expert knowledge. Taste parameters learned by the neural network are fed into the choice model and link the two modules. Our approach extends the L-MNL model (Sifringer et al., 2020) by allowing the neural network to learn the interactions between individual characteristics and alternative attributes. Moreover, we formalize and strengthen the interpretability condition - requiring realistic estimates of behavior indicators (e.g., value-of-time, elasticity) at the disaggregated level, which is crucial for a model to be suitable for scenario analysis and policy decisions. Through a unique network architecture and parameter transformation, we incorporate prior knowledge and guide the neural network to output realistic behavior indicators at the disaggregated level. We show that TasteNet-MNL reaches the ground-truth model's predictability and recovers the nonlinear taste functions on synthetic data. Its estimated value-of-time and choice elasticities at the individual level are close to the ground truth. On a publicly available Swissmetro dataset, TasteNet-MNL outperforms benchmarking MNLs and Mixed Logit model's predictability. It learns a broader spectrum of taste variations within the population and suggests a higher average value-of-time.

econ.EM↗

arXiv2022

Open vs Closed-ended questions in attitudinal surveys -- comparing, combining, and interpreting using natural language processing

To improve the traveling experience, researchers have been analyzing the role of attitudes in travel behavior modeling. Although most researchers use closed-ended surveys, the appropriate method to measure attitudes is debatable. Topic Modeling could significantly reduce the time to extract information from open-ended responses and eliminate subjective bias, thereby alleviating analyst concerns. Our research uses Topic Modeling to extract information from open-ended questions and compare its performance with closed-ended responses. Furthermore, some respondents might prefer answering questions using their preferred questionnaire type. So, we propose a modeling framework that allows respondents to use their preferred questionnaire type to answer the survey and enable analysts to use the modeling frameworks of their choice to predict behavior. We demonstrate this using a dataset collected from the USA that measures the intention to use Autonomous Vehicles for commute trips. Respondents were presented with alternative questionnaire versions (open- and closed- ended). Since our objective was also to compare the performance of alternative questionnaire versions, the survey was designed to eliminate influences resulting from statements, behavioral framework, and the choice experiment. Results indicate the suitability of using Topic Modeling to extract information from open-ended responses; however, the models estimated using the closed-ended questions perform better compared to them. Besides, the proposed model performs better compared to the models used currently. Furthermore, our proposed framework will allow respondents to choose the questionnaire type to answer, which could be particularly beneficial to them when using voice-based surveys.

econ.GN↗

arXiv2022

Transfer learning for cross-modal demand prediction of bike-share and public transit

The urban transportation system is a combination of multiple transport modes, and the interdependencies across those modes exist. This means that the travel demand across different travel modes could be correlated as one mode may receive demand from or create demand for another mode, not to mention natural correlations between different demand time series due to general demand flow patterns across the network. It is expectable that cross-modal ripple effects become more prevalent, with Mobility as a Service. Therefore, by propagating demand data across modes, a better demand prediction could be obtained. To this end, this study explores various machine learning models and transfer learning strategies for cross-modal demand prediction. The trip data of bike-share, metro, and taxi are processed as the station-level passenger flows, and then the proposed prediction method is tested in the large-scale case studies of Nanjing and Chicago. The results suggest that prediction models with transfer learning perform better than unimodal prediction models. Furthermore, stacked Long Short-Term Memory model performs particularly well in cross-modal demand prediction. These results verify our combined method's forecasting improvement over existing benchmarks and demonstrate the good transferability for cross-modal demand prediction in multiple cities.

cs.LG↗

arXiv2021

Predictive and Prescriptive Performance of Bike-Sharing Demand Forecasts for Inventory Management

Bike-sharing systems are a rapidly developing mode of transportation and provide an efficient alternative to passive, motorized personal mobility. The asymmetric nature of bike demand causes the need for rebalancing bike stations, which is typically done during night time. To determine the optimal starting inventory level of a station for a given day, a User Dissatisfaction Function (UDF) models user pickups and returns as non-homogeneous Poisson processes with piece-wise linear rates. In this paper, we devise a deep generative model directly applicable in the UDF by introducing a variational Poisson recurrent neural network model (VP-RNN) to forecast future pickup and return rates. We empirically evaluate our approach against both traditional and learning-based forecasting methods on real trip travel data from the city of New York, USA, and show how our model outperforms benchmarks in terms of system efficiency and demand satisfaction. By explicitly focusing on the combination of decision-making algorithms with learning-based forecasting methods, we highlight a number of shortcomings in literature. Crucially, we show how more accurate predictions do not necessarily translate into better inventory decisions. By providing insights into the interplay between forecasts, model assumptions, and decisions, we point out that forecasts and decision models should be carefully evaluated and harmonized to optimally control shared mobility systems.

math.OC↗

arXiv2020

Semi-nonparametric Latent Class Choice Model with a Flexible Class Membership Component: A Mixture Model Approach

This study presents a semi-nonparametric Latent Class Choice Model (LCCM) with a flexible class membership component. The proposed model formulates the latent classes using mixture models as an alternative approach to the traditional random utility specification with the aim of comparing the two approaches on various measures including prediction accuracy and representation of heterogeneity in the choice process. Mixture models are parametric model-based clustering techniques that have been widely used in areas such as machine learning, data mining and patter recognition for clustering and classification problems. An Expectation-Maximization (EM) algorithm is derived for the estimation of the proposed model. Using two different case studies on travel mode choice behavior, the proposed model is compared to traditional discrete choice models on the basis of parameter estimates' signs, value of time, statistical goodness-of-fit measures, and cross-validation tests. Results show that mixture models improve the overall performance of latent class choice models by providing better out-of-sample prediction accuracy in addition to better representations of heterogeneity without weakening the behavioral and economic interpretability of the choice models.

econ.EM↗

arXiv2019

Multi-output Bus Travel Time Prediction with Convolutional LSTM Neural Network

Accurate and reliable travel time predictions in public transport networks are essential for delivering an attractive service that is able to compete with other modes of transport in urban areas. The traditional application of this information, where arrival and departure predictions are displayed on digital boards, is highly visible in the city landscape of most modern metropolises. More recently, the same information has become critical as input for smart-phone trip planners in order to alert passengers about unreachable connections, alternative route choices and prolonged travel times. More sophisticated Intelligent Transport Systems (ITS) include the predictions of connection assurance, i.e. to hold back services in case a connecting service is delayed. In order to operate such systems, and to ensure the confidence of passengers in the systems, the information provided must be accurate and reliable. Traditional methods have trouble with this as congestion, and thus travel time variability, increases in cities, consequently making travel time predictions in urban areas a non-trivial task. This paper presents a system for bus travel time prediction that leverages the non-static spatio-temporal correlations present in urban bus networks, allowing the discovery of complex patterns not captured by traditional methods. The underlying model is a multi-output, multi-time-step, deep neural network that uses a combination of convolutional and long short-term memory (LSTM) layers. The method is empirically evaluated and compared to other popular approaches for link travel time prediction and currently available services, including the currently deployed model in Copenhagen, Denmark. We find that the proposed model significantly outperforms all the other methods we compare with, and is able to detect small irregular peaks in bus travel times very quickly.

stat.ML↗