-
Applied metamodelling for ATM performance simulations
Authors:
Christoffer Riis,
Francisco N. Antunes,
Tatjana Bolić,
Gérald Gurtner,
Andrew Cook,
Carlos Lima Azevedo,
Francisco Câmara Pereira
Abstract:
The use of Air traffic management (ATM) simulators for planing and operations can be challenging due to their modelling complexity. This paper presents XALM (eXplainable Active Learning Metamodel), a three-step framework integrating active learning and SHAP (SHapley Additive exPlanations) values into simulation metamodels for supporting ATM decision-making. XALM efficiently uncovers hidden relatio…
▽ More
The use of Air traffic management (ATM) simulators for planing and operations can be challenging due to their modelling complexity. This paper presents XALM (eXplainable Active Learning Metamodel), a three-step framework integrating active learning and SHAP (SHapley Additive exPlanations) values into simulation metamodels for supporting ATM decision-making. XALM efficiently uncovers hidden relationships among input and output variables in ATM simulators, those usually of interest in policy analysis. Our experiments show XALM's predictive performance comparable to the XGBoost metamodel with fewer simulations. Additionally, XALM exhibits superior explanatory capabilities compared to non-active learning metamodels.
Using the `Mercury' (flight and passenger) ATM simulator, XALM is applied to a real-world scenario in Paris Charles de Gaulle airport, extending an arrival manager's range and scope by analysing six variables. This case study illustrates XALM's effectiveness in enhancing simulation interpretability and understanding variable interactions. By addressing computational challenges and improving explainability, XALM complements traditional simulation-based analyses.
Lastly, we discuss two practical approaches for reducing the computational burden of the metamodelling further: we introduce a stop** criterion for active learning based on the inherent uncertainty of the metamodel, and we show how the simulations used for the metamodel can be reused across key performance indicators, thus decreasing the overall number of simulations needed.
△ Less
Submitted 7 August, 2023;
originally announced August 2023.
-
Does pre-training on brain-related tasks results in better deep-learning-based brain age biomarkers?
Authors:
Bruno Machado Pacheco,
Victor Hugo Rocha de Oliveira,
Augusto Braga Fernandes Antunes,
Saulo Domingos de Souza Pedro,
Danilo Silva
Abstract:
Brain age prediction using neuroimaging data has shown great potential as an indicator of overall brain health and successful aging, as well as a disease biomarker. Deep learning models have been established as reliable and efficient brain age estimators, being trained to predict the chronological age of healthy subjects. In this paper, we investigate the impact of a pre-training step on deep lear…
▽ More
Brain age prediction using neuroimaging data has shown great potential as an indicator of overall brain health and successful aging, as well as a disease biomarker. Deep learning models have been established as reliable and efficient brain age estimators, being trained to predict the chronological age of healthy subjects. In this paper, we investigate the impact of a pre-training step on deep learning models for brain age prediction. More precisely, instead of the common approach of pre-training on natural imaging classification, we propose pre-training the models on brain-related tasks, which led to state-of-the-art results in our experiments on ADNI data. Furthermore, we validate the resulting brain age biomarker on images of patients with mild cognitive impairment and Alzheimer's disease. Interestingly, our results indicate that better-performing deep learning models in terms of brain age prediction on healthy patients do not result in more reliable biomarkers.
△ Less
Submitted 11 July, 2023;
originally announced July 2023.
-
NOSTROMO: Lessons learned, conclusions and way forward
Authors:
Mayte Cano,
Andrés Perillo,
Juan Antonio López,
Faustino Tello,
Javier Poveda,
Francisco Câmara,
Francisco Antunes,
Christoffer Riis,
Ian Crook,
Abderrazak Tibichte,
Sandrine Molton,
David Mocholí,
Ricardo Herranz,
Gérald Gurtner,
Tatjana Bolić,
Andrew Cook,
Jovana Kuljanin,
Xavier Prats
Abstract:
This White Paper sets out to explain the value that metamodelling can bring to air traffic management (ATM) research. It will define metamodelling and explore what it can, and cannot, do. The reader is assumed to have basic knowledge of SESAR: the Single European Sky ATM Research project. An important element of SESAR, as the technological pillar of the Single European Sky initiative, is to bring…
▽ More
This White Paper sets out to explain the value that metamodelling can bring to air traffic management (ATM) research. It will define metamodelling and explore what it can, and cannot, do. The reader is assumed to have basic knowledge of SESAR: the Single European Sky ATM Research project. An important element of SESAR, as the technological pillar of the Single European Sky initiative, is to bring about improvements, as measured through specific key performance indicators (KPIs), and as implemented by a series of so-called SESAR 'Solutions'. These 'Solutions' are new or improved operational procedures or technologies, designed to meet operational and performance improvements described in the European ATM Master Plan.
△ Less
Submitted 29 March, 2023;
originally announced March 2023.
-
Bayesian Active Learning with Fully Bayesian Gaussian Processes
Authors:
Christoffer Riis,
Francisco Antunes,
Frederik Boe Hüttel,
Carlos Lima Azevedo,
Francisco Câmara Pereira
Abstract:
The bias-variance trade-off is a well-known problem in machine learning that only gets more pronounced the less available data there is. In active learning, where labeled data is scarce or difficult to obtain, neglecting this trade-off can cause inefficient and non-optimal querying, leading to unnecessary data labeling. In this paper, we focus on active learning with Gaussian Processes (GPs). For…
▽ More
The bias-variance trade-off is a well-known problem in machine learning that only gets more pronounced the less available data there is. In active learning, where labeled data is scarce or difficult to obtain, neglecting this trade-off can cause inefficient and non-optimal querying, leading to unnecessary data labeling. In this paper, we focus on active learning with Gaussian Processes (GPs). For the GP, the bias-variance trade-off is made by optimization of the two hyperparameters: the length scale and noise-term. Considering that the optimal mode of the joint posterior of the hyperparameters is equivalent to the optimal bias-variance trade-off, we approximate this joint posterior and utilize it to design two new acquisition functions. The first one is a Bayesian variant of Query-by-Committee (B-QBC), and the second is an extension that explicitly minimizes the predictive variance through a Query by Mixture of Gaussian Processes (QB-MGP) formulation. Across six simulators, we empirically show that B-QBC, on average, achieves the best marginal likelihood, whereas QB-MGP achieves the best predictive performance. We show that incorporating the bias-variance trade-off in the acquisition functions mitigates unnecessary and expensive data labeling.
△ Less
Submitted 14 January, 2023; v1 submitted 20 May, 2022;
originally announced May 2022.
-
Modeling and control of malaria dynamics in fish farming regions
Authors:
Felipe J. P. Antunes,
M. Soledad Aronna,
Cláudia T. Codeço
Abstract:
In this work we propose a model that represents the relation between fish ponds, the mosquito population and the transmission of malaria.
It has been observed that in the Amazonic region of Acre, in the North of Brazil, fish farming is correlated to the transmission of malaria when carried out in artificial ponds that become breeding sites. Evidence has been found indicating that cleaning the ve…
▽ More
In this work we propose a model that represents the relation between fish ponds, the mosquito population and the transmission of malaria.
It has been observed that in the Amazonic region of Acre, in the North of Brazil, fish farming is correlated to the transmission of malaria when carried out in artificial ponds that become breeding sites. Evidence has been found indicating that cleaning the vegetation from the edges of the crop tanks helps to control the size of the mosquito population.
We use our model to determine the effective contribution of fish farming practices on malaria transmission dynamics. The model consists of a nonlinear system of ordinary differential equations with jumps at the cleaning time, which act as impulsive controls. We study the asymptotic behaviour of the system in function of the intensity and periodicity of the cleaning, and the value of the parameters. In particular, we state sufficient conditions under which the mosquito population is eliminated or persists, and under which the malaria is eliminated or becomes endemic. We prove our conditions by applying results for cooperative systems with concave nonlinearities.
△ Less
Submitted 26 April, 2023; v1 submitted 27 October, 2020;
originally announced October 2020.
-
Evidence of Fraud in Brazil's Electoral Campaigns Via the Benford's Law
Authors:
Daniel Gamermann,
Felipe Leite Antunes
Abstract:
The principle of democracy is that the people govern through elected representatives. Therefore, a democracy is healthy as long as the elected politicians do represent the people. We have analyzed data from the Brazilian electoral court (Tribunal Superior Eleitoral, TSE) concerning money donations for the electoral campaigns and the election results. Our work points to two disturbing conclusions:…
▽ More
The principle of democracy is that the people govern through elected representatives. Therefore, a democracy is healthy as long as the elected politicians do represent the people. We have analyzed data from the Brazilian electoral court (Tribunal Superior Eleitoral, TSE) concerning money donations for the electoral campaigns and the election results. Our work points to two disturbing conclusions: money is a determining factor on whether a candidate is elected or not (as opposed to representativeness); secondly, the use of Benford's Law to analyze the declared donations received by the parties and electoral campaigns shows evidence of fraud in the declarations. A better term to define Brazil's government system is what we define as chrimatocracy (govern by money).
△ Less
Submitted 11 July, 2017;
originally announced July 2017.
