Showing 1–15 of 15 results for author: Toledo, R

With Friends Like These, Who Needs Explanations? Evaluating User Understanding of Group Recommendations

Authors: Cedric Waterschoot, Raciel Yera Toledo, Nava Tintarev, Francesco Barile

Abstract: Group Recommender Systems (GRS) employing social choice-based aggregation strategies have previously been explored in terms of perceived consensus, fairness, and satisfaction. At the same time, the impact of textual explanations has been examined, but the results suggest a low effectiveness of these explanations. However, user understanding remains fairly unexplored, even if it can contribute posi… ▽ More Group Recommender Systems (GRS) employing social choice-based aggregation strategies have previously been explored in terms of perceived consensus, fairness, and satisfaction. At the same time, the impact of textual explanations has been examined, but the results suggest a low effectiveness of these explanations. However, user understanding remains fairly unexplored, even if it can contribute positively to transparent GRS. This is particularly interesting to study in more complex or potentially unfair scenarios when user preferences diverge, such as in a minority scenario (where group members have similar preferences, except for a single member in a minority position). In this paper, we analyzed the impact of different types of explanations on user understanding of group recommendations. We present a randomized controlled trial (n = 271) using two between-subject factors: (i) the aggregation strategy (additive, least misery, and approval voting), and (ii) the modality of explanation (no explanation, textual explanation, or multimodal explanation). We measured both subjective (self-perceived by the user) and objective understanding (performance on model simulation, counterfactuals and error detection). In line with recent findings on explanations for machine learning models, our results indicate that more detailed explanations, whether textual or multimodal, did not increase subjective or objective understanding. However, we did find a significant effect of aggregation strategies on both subjective and objective understanding. These results imply that when constructing GRS, practitioners need to consider that the choice of aggregation strategy can influence the understanding of users. Post-hoc analysis also suggests that there is value in analyzing performance on different tasks, rather than through a single aggregated metric of understanding. △ Less

Submitted 7 May, 2025; originally announced May 2025.

Comments: This article will be part of the UMAP 2025 conference. (33rd ACM Conference on User Modeling, Adaptation and Personalization (UMAP25), June 16-19, 2025, New York City, NY, USA)

A Performance Increment Strategy for Semantic Segmentation of Low-Resolution Images from Damaged Roads

Authors: Rafael S. Toledo, Cristiano S. Oliveira, Vitor H. T. Oliveira, Eric A. Antonelo, Aldo von Wangenheim

Abstract: Autonomous driving needs good roads, but 85% of Brazilian roads have damages that deep learning models may not regard as most semantic segmentation datasets for autonomous driving are high-resolution images of well-maintained urban roads. A representative dataset for emerging countries consists of low-resolution images of poorly maintained roads and includes labels of damage classes; in this scena… ▽ More Autonomous driving needs good roads, but 85% of Brazilian roads have damages that deep learning models may not regard as most semantic segmentation datasets for autonomous driving are high-resolution images of well-maintained urban roads. A representative dataset for emerging countries consists of low-resolution images of poorly maintained roads and includes labels of damage classes; in this scenario, three challenges arise: objects with few pixels, objects with undefined shapes, and highly underrepresented classes. To tackle these challenges, this work proposes the Performance Increment Strategy for Semantic Segmentation (PISSS) as a methodology of 14 training experiments to boost performance. With PISSS, we reached state-of-the-art results of 79.8 and 68.8 mIoU on the Road Traversing Knowledge (RTK) and Technik Autonomer Systeme 500 (TAS500) test sets, respectively. Furthermore, we also offer an analysis of DeepLabV3+ pitfalls for small object segmentation. △ Less

Submitted 25 November, 2024; originally announced November 2024.

Comprehending Variability in Analysis Results of Software Product Lines

Authors: Rafael F. Toledo, Joanne M. Atlee, Rui Ming Xiong

Abstract: Analyses of a software product line (SPL) typically report variable results that are annotated with logical expressions indicating the set of product variants for which the results hold. These expressions can get complicated and difficult to reason about when the SPL has lots of features and product variants. Previous work introduced a visualizer that supports filters for highlighting the analysis… ▽ More Analyses of a software product line (SPL) typically report variable results that are annotated with logical expressions indicating the set of product variants for which the results hold. These expressions can get complicated and difficult to reason about when the SPL has lots of features and product variants. Previous work introduced a visualizer that supports filters for highlighting the analysis results that apply to product variants of interest, but this work was weakly evaluated. In this paper, we report on a controlled user study that evaluates the effectiveness of this new visualizer in helping the user search variable results and compare the results of multiple variants. Our findings indicate that the use of the new visualizer significantly improves the correctness and efficiency of the user's work and reduces the user's cognitive load in working with variable results. △ Less

Submitted 30 October, 2023; originally announced October 2023.

Fluctuating Sunspot Numbers Exhibit A Non-Markovian Damped Stochastic Process

Authors: Reynan L. Toledo, Reinabelle Reyes, Christopher C. Bernido

Abstract: The rise and fall in the number of sunspots have served as a lynchpin in many investigations on solar dynamics. Arising from magnetic disturbances in the sun, variations in sunspot numbers have helped define a solar cycle of around eleven years which to date is yet to be fully understood. We model the fluctuation of sunspot numbers as a modulated Brownian motion characterized by a memory parameter… ▽ More The rise and fall in the number of sunspots have served as a lynchpin in many investigations on solar dynamics. Arising from magnetic disturbances in the sun, variations in sunspot numbers have helped define a solar cycle of around eleven years which to date is yet to be fully understood. We model the fluctuation of sunspot numbers as a modulated Brownian motion characterized by a memory parameter μ and a decay parameter \b{eta}. By matching the theoretical and empirical mean square deviation of the sunspot numbers, the values of μ and \b{eta} are determined for each solar cycle. This allows us to obtain an exact form of a probability density function (PDF) which closely matches the dataset for sunspots. This novel PDF for sunspot numbers exhibit a memory behavior from which some insights could be obtained. In particular, the values of μ indicate that consecutive sunspot numbers are negatively correlated for large times. The values of \b{eta}, on the other hand, when viewed as a time series from one solar cycle to another, indicate a positive trend towards increasing values which could possibly suggest a diminishing solar activity. △ Less

Submitted 25 June, 2023; originally announced June 2023.

Comments: 22 pages, 12 Figures

MSC Class: 60

Face Reconstruction with Variational Autoencoder and Face Masks

Authors: Rafael S. Toledo, Eric A. Antonelo

Abstract: Variational AutoEncoders (VAE) employ deep learning models to learn a continuous latent z-space that is subjacent to a high-dimensional observed dataset. With that, many tasks are made possible, including face reconstruction and face synthesis. In this work, we investigated how face masks can help the training of VAEs for face reconstruction, by restricting the learning to the pixels selected by t… ▽ More Variational AutoEncoders (VAE) employ deep learning models to learn a continuous latent z-space that is subjacent to a high-dimensional observed dataset. With that, many tasks are made possible, including face reconstruction and face synthesis. In this work, we investigated how face masks can help the training of VAEs for face reconstruction, by restricting the learning to the pixels selected by the face mask. An evaluation of the proposal using the celebA dataset shows that the reconstructed images are enhanced with the face masks, especially when SSIM loss is used either with l1 or l2 loss functions. We noticed that the inclusion of a decoder for face mask prediction in the architecture affected the performance for l1 or l2 loss functions, while this was not the case for the SSIM loss. Besides, SSIM perceptual loss yielded the crispest samples between all hypotheses tested, although it shifts the original color of the image, making the usage of the l1 or l2 losses together with SSIM helpful to solve this issue. △ Less

Submitted 3 December, 2021; originally announced December 2021.

Comments: 12 pages, 7 figures, 18th Encontro Nacional de Inteligência Artificial e Computacional (ENIAC)

Applying Declarative Analysis to Software Product Line Models: An Industrial Study

Authors: Ramy Shahin, Robert Hackman, Rafael Toledo, Ramesh S, Joanne M. Atlee, Marsha Chechik

Abstract: Software Product Lines (SPLs) are families of related software products developed from a common set of artifacts. Most existing analysis tools can be applied to a single product at a time, but not to an entire SPL. Some tools have been redesigned/re-implemented to support the kind of variability exhibited in SPLs, but this usually takes a lot of effort, and is error-prone. Declarative analyses wri… ▽ More Software Product Lines (SPLs) are families of related software products developed from a common set of artifacts. Most existing analysis tools can be applied to a single product at a time, but not to an entire SPL. Some tools have been redesigned/re-implemented to support the kind of variability exhibited in SPLs, but this usually takes a lot of effort, and is error-prone. Declarative analyses written in languages like Datalog have been collectively lifted to SPLs in prior work, which makes the process of applying an existing declarative analysis to a product line more straightforward. In this paper, we take an existing declarative analysis (behaviour alteration) written in the Grok declarative language, port it to Datalog, and apply it to a set of automotive software product lines from General Motors. We discuss the design of the analysis pipeline used in this process, present its scalability results, and provide a means to visualize the analysis results for a subset of products filtered by feature expression. We also reflect on some of the lessons learned throughout this project. △ Less

Submitted 30 July, 2021; v1 submitted 15 July, 2021; originally announced July 2021.

Comments: MODELS'21 pre-print

Electron Paramagnetic Resonance signature of point defects in neutron irradiated hexagonal Boron Nitride

Authors: J. R. Toledo, D. B. de Jesus, M. Kianinia, A. S. Leal, C. Fantini, L. A. Cury, G. M. Sáfar, I. Aharonovich, K. Krambrock

Abstract: Hexagonal boron nitride (h-BN) is an attractive van der Waals material for studying fluorescent defects due to its large bandgap. In this work, we demonstrate enhanced pink color due to neutron irradiation and perform electron paramagnetic resonance (EPR) measurements. The new point defects are tentatively assigned to doubly- occupied nitrogen vacancies with (S = 1) and a zero-field splitting (D =… ▽ More Hexagonal boron nitride (h-BN) is an attractive van der Waals material for studying fluorescent defects due to its large bandgap. In this work, we demonstrate enhanced pink color due to neutron irradiation and perform electron paramagnetic resonance (EPR) measurements. The new point defects are tentatively assigned to doubly- occupied nitrogen vacancies with (S = 1) and a zero-field splitting (D = 1.2 GHz). These defects are associated with a broad visible optical absorption band and near infrared photoluminescence band centered at ~ 490 nm and 820 nm, respectively. The EPR signal intensities are strongly affected by thermal treatments in temperature range between 600 to 800°C, where also the irradiation - induced pink color is lost. Our results are important for understanding of point defects in h-BN and their deployment for quantum and integrated photonic applications. △ Less

Submitted 8 October, 2018; v1 submitted 14 July, 2018; originally announced July 2018.

Journal ref: Phys. Rev. B 98, 155203 (2018)

Mix-ORAM: Using delegate shuffles

Authors: Raphael R. Toledo, George Danezis, Isao Echizen

Abstract: Oblivious RAM (ORAM) is a key technology for providing private storage and querying on untrusted machines but is commonly seen as impractical due to the high overhead of the re-randomization, called the eviction, the client incurs. We propose in this work to securely delegate the eviction to semi-trusted third parties to enable any client to accede the ORAM technology and present four different de… ▽ More Oblivious RAM (ORAM) is a key technology for providing private storage and querying on untrusted machines but is commonly seen as impractical due to the high overhead of the re-randomization, called the eviction, the client incurs. We propose in this work to securely delegate the eviction to semi-trusted third parties to enable any client to accede the ORAM technology and present four different designs inspired by mix-net technologies with reasonable periodic costs. △ Less

Submitted 4 September, 2017; originally announced September 2017.

Lower-Cost epsilon-Private Information Retrieval

Authors: Raphael R. Toledo, George Danezis, Ian Goldberg

Abstract: Private Information Retrieval (PIR), despite being well studied, is computationally costly and arduous to scale. We explore lower-cost relaxations of information-theoretic PIR, based on dummy queries, sparse vectors, and compositions with an anonymity system. We prove the security of each scheme using a flexible differentially private definition for private queries that can capture notions of impe… ▽ More Private Information Retrieval (PIR), despite being well studied, is computationally costly and arduous to scale. We explore lower-cost relaxations of information-theoretic PIR, based on dummy queries, sparse vectors, and compositions with an anonymity system. We prove the security of each scheme using a flexible differentially private definition for private queries that can capture notions of imperfect privacy. We show that basic schemes are weak, but some of them can be made arbitrarily safe by composing them with large anonymity systems. △ Less

Submitted 1 April, 2016; originally announced April 2016.

TDD for Embedded Systems: A Basic Approach and Toolset

Authors: Rogerio Atem de Carvalho, Hudson Silva, Rafael Ferreira Toledo, Milena Silveira de Azevedo

Abstract: The evolution of information technology and electronics in general has been consistently increasing the use of embedded systems. While hardware development for these systems is already consistent, software development for embedded systems still lacks a consolidated methodology. This paper describes a process and toolset for Embedded Systems Validation and Verification using FSM (Finite State Machi… ▽ More The evolution of information technology and electronics in general has been consistently increasing the use of embedded systems. While hardware development for these systems is already consistent, software development for embedded systems still lacks a consolidated methodology. This paper describes a process and toolset for Embedded Systems Validation and Verification using FSM (Finite State Machines) and TDD (Test Driven Development). △ Less

Submitted 3 August, 2015; v1 submitted 28 July, 2015; originally announced July 2015.

Comments: 07 pages, 05 figures, work in progress

Strong convergence theorem of Cesàro means with respect to the Walsh system

Authors: István Blahota, George Tephnadze, Rodolfo Toledo

Abstract: We prove that Cesàro means of one-dimensional Walsh-Fourier series are uniformly bounded operators in the martingale Hardy space $H_{p}$ for $% 0<p<1/\left( 1+α\right).$ We prove that Cesàro means of one-dimensional Walsh-Fourier series are uniformly bounded operators in the martingale Hardy space $H_{p}$ for $% 0<p<1/\left( 1+α\right).$ △ Less

Submitted 21 April, 2015; originally announced April 2015.

Comments: arXiv admin note: substantial text overlap with arXiv:1504.05778

MSC Class: 42C10

Microemulsification: An Approach for Analytical Determinations

Authors: Renato S. Lima, Leandro Y. Shiroma, Alvaro V. N. C. Teixeira, José R. de Toledo, Bruno C. do Couto, Rogério M. de Carvalho, Emanuel Carrilho, Lauro T. Kubota, Angelo L. Gobbi

Abstract: We address a novel method for analytical determinations that combines simplicity, rapidity, low consumption of chemicals, and portability with high analytical performance taking into account parameters such as precision, linearity, robustness, and accuracy. This approach relies on the effect of the analyte content over the Gibbs free energy of dispersions, affecting the thermodynamic stabilization… ▽ More We address a novel method for analytical determinations that combines simplicity, rapidity, low consumption of chemicals, and portability with high analytical performance taking into account parameters such as precision, linearity, robustness, and accuracy. This approach relies on the effect of the analyte content over the Gibbs free energy of dispersions, affecting the thermodynamic stabilization of emulsions or Winsor systems to form microemulsions (MEs). Such phenomenon was expressed by the minimum volume fraction of amphiphile required to form microemulsion, which was the analytical signal of the method. The performed studies were: phase behavior, droplet dimension by dynamic light scattering, analytical curve, and robustness tests. The reliability of the method was evaluated by determining water in ethanol fuels and monoethylene glycol in complex samples of liquefied natural gas. The dispersions were composed of water-chlorobenzene (water analysis) and water-oleic acid (monoethylene glycol analysis) with ethanol as the hydrotrope phase. The experiments to determine water demonstrated that the analytical performance depends on the composition of ME. The linear range was fairly broad with limits of linearity up to 70.00% water in ethanol. For monoethylene glycol in water the linear range was observed throughout the volume fraction of analyte. The natural gas samples provided by the Petrobras exhibited color, particulate material, high ionic strength, and diverse compounds as metals, carboxylic acids, and anions. The method allowed accurate measures bypassing steps such as extraction, preconcentration, and dilution of the sample. In addition, the levels of robustness were promising. This parameter was evaluated by investigating the effect of (i) deviations in volumetric preparation of the dispersions and (ii) changes in temperature over the analyte contents recorded by the method. △ Less

Submitted 8 September, 2014; originally announced September 2014.

Comments: 9 pages (paper) + 12 pages (Supplementary Information included), 7 figures (paper) + 6 figures (S. I.)

Journal ref: Anal. Chem., Article ASAP (2014)

La optica de imagenes en la extension universitaria de Unicamp

Authors: Jose Joaquin Lunazzi, Daniel Souza Ferreira Magalhaes, Maria Clara Igrejas Amon, Rolando Serra Toledo

Abstract: The physics teaching particularly the optics teaching always have faced motivational problems, mostly due to the loss of the link between the social practice and the daily life of the student. Moreover, the laboratory practices have lost space in high school attributable to a lack of resources or caused by the fact that the main purpose of the high school is being no more than the access to the Un… ▽ More The physics teaching particularly the optics teaching always have faced motivational problems, mostly due to the loss of the link between the social practice and the daily life of the student. Moreover, the laboratory practices have lost space in high school attributable to a lack of resources or caused by the fact that the main purpose of the high school is being no more than the access to the University. This work shows how an extension activity called "Holographic Exhibit" tries to motivate high school students by means of experimentation, observation and the application of physics concepts in daily life, awakening their interests to the general physics. La ensenanza de la fisica y en particular de la optica siempre ha enfrentado problemas de aspecto motivacional principalmente por la perdida del vinculo con la practica social y con la vida cotidiana del estudiante. Ademas, las practicas de laboratorio han perdido mucho espacio en la ensenanza media por carencias de las escuelas o debido a que tiene apenas como objeto el examen de ingreso a la universidad. El objetivo de este trabajo es mostrar como una actividad de extensión universitaria llamada 'Exposicion de Holografia' intenta motivar a estudiantes de enseñanza media a traves de la experimentacion, observacion y aplicacion de conceptos fisicos en la vida cotidiana, despertando sus intereses por la física general. △ Less

Submitted 30 December, 2012; originally announced December 2012.

Comments: 8 pages, 8 figures. in Portuguese

Journal ref: Latinoamerican Journal of Physics Education, Vol. 4, No. 1, Jan. 2010, p 441-448

Euclid Definition Study Report

Authors: R. Laureijs, J. Amiaux, S. Arduini, J. -L. Auguères, J. Brinchmann, R. Cole, M. Cropper, C. Dabin, L. Duvet, A. Ealet, B. Garilli, P. Gondoin, L. Guzzo, J. Hoar, H. Hoekstra, R. Holmes, T. Kitching, T. Maciaszek, Y. Mellier, F. Pasian, W. Percival, J. Rhodes, G. Saavedra Criado, M. Sauvage, R. Scaramella , et al. (194 additional authors not shown)

Abstract: Euclid is a space-based survey mission from the European Space Agency designed to understand the origin of the Universe's accelerating expansion. It will use cosmological probes to investigate the nature of dark energy, dark matter and gravity by tracking their observational signatures on the geometry of the universe and on the cosmic history of structure formation. The mission is optimised for tw… ▽ More Euclid is a space-based survey mission from the European Space Agency designed to understand the origin of the Universe's accelerating expansion. It will use cosmological probes to investigate the nature of dark energy, dark matter and gravity by tracking their observational signatures on the geometry of the universe and on the cosmic history of structure formation. The mission is optimised for two independent primary cosmological probes: Weak gravitational Lensing (WL) and Baryonic Acoustic Oscillations (BAO). The Euclid payload consists of a 1.2 m Korsch telescope designed to provide a large field of view. It carries two instruments with a common field-of-view of ~0.54 deg2: the visual imager (VIS) and the near infrared instrument (NISP) which contains a slitless spectrometer and a three bands photometer. The Euclid wide survey will cover 15,000 deg2 of the extragalactic sky and is complemented by two 20 deg2 deep fields. For WL, Euclid measures the shapes of 30-40 resolved galaxies per arcmin2 in one broad visible R+I+Z band (550-920 nm). The photometric redshifts for these galaxies reach a precision of dz/(1+z) < 0.05. They are derived from three additional Euclid NIR bands (Y, J, H in the range 0.92-2.0 micron), complemented by ground based photometry in visible bands derived from public data or through engaged collaborations. The BAO are determined from a spectroscopic survey with a redshift accuracy dz/(1+z) =0.001. The slitless spectrometer, with spectral resolution ~250, predominantly detects Ha emission line galaxies. Euclid is a Medium Class mission of the ESA Cosmic Vision 2015-2025 programme, with a foreseen launch date in 2019. This report (also known as the Euclid Red Book) describes the outcome of the Phase A study. △ Less

Submitted 14 October, 2011; originally announced October 2011.

Comments: 116 pages, with executive summary and table of contents

Report number: ESA/SRE(2011)12

Evaluation of Three Vision Based Object Perception Methods for a Mobile Robot

Authors: Arnau Ramisa, David Aldavert, Shrihari Vasudevan, Ricardo Toledo, Ramon Lopez de Mantaras

Abstract: This paper addresses object perception applied to mobile robotics. Being able to perceive semantically meaningful objects in unstructured environments is a key capability in order to make robots suitable to perform high-level tasks in home environments. However, finding a solution for this task is daunting: it requires the ability to handle the variability in image formation in a moving camera wit… ▽ More This paper addresses object perception applied to mobile robotics. Being able to perceive semantically meaningful objects in unstructured environments is a key capability in order to make robots suitable to perform high-level tasks in home environments. However, finding a solution for this task is daunting: it requires the ability to handle the variability in image formation in a moving camera with tight time constraints. The paper brings to attention some of the issues with applying three state of the art object recognition and detection methods in a mobile robotics scenario, and proposes methods to deal with windowing/segmentation. Thus, this work aims at evaluating the state-of-the-art in object perception in an attempt to develop a lightweight solution for mobile robotics use/research in typical indoor settings. △ Less

Submitted 2 February, 2011; originally announced February 2011.

Comments: 37 pages, 11 figures

Report number: IIIA research report 2011-01