-
Stacked conformal prediction
Authors:
Paulo C. Marques F
Abstract:
We consider the conformalization of a stacked ensemble of predictive models, showing that the potentially simple form of the meta-learner at the top of the stack enables a procedure with manageable computational cost that achieves approximate marginal validity without requiring the use of a separate calibration sample. Empirical results indicate that the method compares favorably to a standard ind…
▽ More
We consider the conformalization of a stacked ensemble of predictive models, showing that the potentially simple form of the meta-learner at the top of the stack enables a procedure with manageable computational cost that achieves approximate marginal validity without requiring the use of a separate calibration sample. Empirical results indicate that the method compares favorably to a standard inductive alternative.
△ Less
Submitted 18 May, 2025;
originally announced May 2025.
-
A new architecture for high speed core-selective switch for multicore fibers
Authors:
Cristóbal Melo,
Matías Reyes. F.,
Diego Arroyo,
Esteban S. Gómez,
Stephen P. Walborn,
Gustavo Lima,
Miguel Figueroa,
Jaime Cariñe,
Gabriel Saavedra
Abstract:
The use of multicore optical fibers is now recognized as one of the most promising methods to implement the space-division multiplexing techniques required to overcome the impending capacity limit of conventional single-mode optical fibers. Nonetheless, new devices for networking operations compatible with these fibers will be required in order to implement the next-generation high-capacity optica…
▽ More
The use of multicore optical fibers is now recognized as one of the most promising methods to implement the space-division multiplexing techniques required to overcome the impending capacity limit of conventional single-mode optical fibers. Nonetheless, new devices for networking operations compatible with these fibers will be required in order to implement the next-generation high-capacity optical networks. In this work, we develop a new architecture to build a high-speed core-selective switch, critical for efficiently distributing signals over the network. The device relies on multicore interference, and can change among outputs in less than 0.7 us, while achieving less than -18 dB of average inter-core crosstalk, making it compatible with a wide range of network switching tasks. The functionality of the device was demonstrated by routing a 1GBs optical signal and by successfully switching signals over a field-installed multicore fiber network. Our results demonstrate for the first time the operation of a multicore optical fiber switch functioning under real-world conditions, with switching speeds that are three orders of magnitude faster than current commercial devices. This new optical switch design is also fully compatible with standard multiplexing techniques and, thus, represents an important achievement towards the integration of high-capacity multicore telecommunication networks.
△ Less
Submitted 26 November, 2024;
originally announced November 2024.
-
Projected random forests and conformal prediction of circular data
Authors:
Paulo C. Marques F.,
Rinaldo Artes,
Helton Graziadei
Abstract:
We apply split conformal prediction techniques to regression problems with circular responses by introducing a suitable conformity score, leading to prediction sets with adaptive arc length and finite-sample coverage guarantees for any circular predictive model under exchangeable data. Leveraging the high performance of existing predictive models designed for linear responses, we analyze a general…
▽ More
We apply split conformal prediction techniques to regression problems with circular responses by introducing a suitable conformity score, leading to prediction sets with adaptive arc length and finite-sample coverage guarantees for any circular predictive model under exchangeable data. Leveraging the high performance of existing predictive models designed for linear responses, we analyze a general projection procedure that converts any linear response regression model into one suitable for circular responses. When random forests serve as basis models in this projection procedure, we harness the out-of-bag dynamics to eliminate the necessity for a separate calibration sample in the construction of prediction sets. For synthetic and real datasets the resulting projected random forests model produces more efficient out-of-bag conformal prediction sets, with shorter median arc length, when compared to the split conformal prediction sets generated by two existing alternative models.
△ Less
Submitted 25 December, 2024; v1 submitted 31 October, 2024;
originally announced October 2024.
-
Conformal prediction for frequency-severity modeling
Authors:
Helton Graziadei,
Paulo C. Marques F.,
Eduardo F. L. de Melo,
Rodrigo S. Targino
Abstract:
We present a model-agnostic framework for the construction of prediction intervals of insurance claims, with finite sample statistical guarantees, extending the technique of split conformal prediction to the domain of two-stage frequency-severity modeling. The framework effectiveness is showcased with simulated and real datasets using classical parametric models and contemporary machine learning m…
▽ More
We present a model-agnostic framework for the construction of prediction intervals of insurance claims, with finite sample statistical guarantees, extending the technique of split conformal prediction to the domain of two-stage frequency-severity modeling. The framework effectiveness is showcased with simulated and real datasets using classical parametric models and contemporary machine learning methods. When the underlying severity model is a random forest, we extend the two-stage split conformal prediction algorithm, showing how the out-of-bag mechanism can be leveraged to eliminate the need for a calibration set in the conformal procedure.
△ Less
Submitted 19 June, 2025; v1 submitted 24 July, 2023;
originally announced July 2023.
-
Universal distribution of the empirical coverage in split conformal prediction
Authors:
Paulo C. Marques F
Abstract:
When split conformal prediction operates in batch mode with exchangeable data, we determine the exact distribution of the empirical coverage of prediction sets produced for a finite batch of future observables, as well as the exact distribution of its almost sure limit when the batch size goes to infinity. Both distributions are universal, being determined solely by the nominal miscoverage level a…
▽ More
When split conformal prediction operates in batch mode with exchangeable data, we determine the exact distribution of the empirical coverage of prediction sets produced for a finite batch of future observables, as well as the exact distribution of its almost sure limit when the batch size goes to infinity. Both distributions are universal, being determined solely by the nominal miscoverage level and the calibration sample size, thereby establishing a criterion for choosing the minimum required calibration sample size in applications.
△ Less
Submitted 21 September, 2024; v1 submitted 5 March, 2023;
originally announced March 2023.
-
Confidence intervals for the random forest generalization error
Authors:
Paulo C. Marques F
Abstract:
We show that the byproducts of the standard training process of a random forest yield not only the well known and almost computationally free out-of-bag point estimate of the model generalization error, but also give a direct path to compute confidence intervals for the generalization error which avoids processes of data splitting and model retraining. Besides the low computational cost involved i…
▽ More
We show that the byproducts of the standard training process of a random forest yield not only the well known and almost computationally free out-of-bag point estimate of the model generalization error, but also give a direct path to compute confidence intervals for the generalization error which avoids processes of data splitting and model retraining. Besides the low computational cost involved in their construction, these confidence intervals are shown through simulations to have good coverage and appropriate shrinking rate of their width in terms of the training sample size.
△ Less
Submitted 11 March, 2022; v1 submitted 11 December, 2021;
originally announced December 2021.
-
Exploiting timing capabilities of the CHEOPS mission with warm-Jupiter planets
Authors:
Borsato L,
Piotto G,
Gandolfi D,
Nascimbeni V,
Lacedelli G,
Marzari F,
Billot N,
Maxted P,
Sousa S G,
Cameron A C,
Bonfanti A,
Wilson T,
Serrano L,
Garai Z,
Alibert Y,
Alonso R,
Asquier J,
Bárczy T,
Bandy T,
Barrado D,
Barros S C,
Baumjohann W,
Beck M,
Beck T,
Benz W
, et al. (53 additional authors not shown)
Abstract:
We present 17 transit light curves of seven known warm-Jupiters observed with the CHaracterising ExOPlanet Satellite (CHEOPS). The light curves have been collected as part of the CHEOPS Guaranteed Time Observation (GTO) program that searches for transit-timing variation (TTV) of warm-Jupiters induced by a possible external perturber to shed light on the evolution path of such planetary systems. We…
▽ More
We present 17 transit light curves of seven known warm-Jupiters observed with the CHaracterising ExOPlanet Satellite (CHEOPS). The light curves have been collected as part of the CHEOPS Guaranteed Time Observation (GTO) program that searches for transit-timing variation (TTV) of warm-Jupiters induced by a possible external perturber to shed light on the evolution path of such planetary systems. We describe the CHEOPS observation process, from the planning to the data analysis. In this work we focused on the timing performance of CHEOPS, the impact of the sampling of the transit phases, and the improvement we can obtain combining multiple transits together. We reached the highest precision on the transit time of about 13-16 s for the brightest target (WASP-38, G = 9.2) in our sample. From the combined analysis of multiple transits of fainter targets with G >= 11 we obtained a timing precision of about 2 min. Additional observations with CHEOPS, covering a longer temporal baseline, will further improve the precision on the transit times and will allow us to detect possible TTV signals induced by an external perturber.
△ Less
Submitted 21 June, 2021;
originally announced June 2021.
-
PARAM: A Microprocessor Hardened for Power Side-Channel Attack Resistance
Authors:
Muhammad Arsath K F,
Vinod Ganesan,
Rahul Bodduna,
Chester Rebeiro
Abstract:
The power consumption of a microprocessor is a huge channel for information leakage. While the most popular exploitation of this channel is to recover cryptographic keys from embedded devices, other applications such as mobile app fingerprinting, reverse engineering of firmware, and password recovery are growing threats. Countermeasures proposed so far are tuned to specific applications, such as c…
▽ More
The power consumption of a microprocessor is a huge channel for information leakage. While the most popular exploitation of this channel is to recover cryptographic keys from embedded devices, other applications such as mobile app fingerprinting, reverse engineering of firmware, and password recovery are growing threats. Countermeasures proposed so far are tuned to specific applications, such as crypto-implementations. They are not scalable to the large number and variety of applications that typically run on a general purpose microprocessor.
In this paper, we investigate the design of a microprocessor, called PARAM with increased resistance to power based side-channel attacks. To design PARAM, we start with identifying the most leaking modules in an open-source RISC V processor. We evaluate the leakage in these modules and then add suitable countermeasures. The countermeasures depend on the cause of leakage in each module and can vary from simple modifications of the HDL code ensuring secure translation by the EDA tools, to obfuscating data and address lines thus breaking correlation with the processor's power consumption. The resultant processor is instantiated on the SASEBO-GIII FPGA board and found to resist Differential Power Analysis even after one million power traces. Compared to contemporary countermeasures for power side-channel attacks, overheads in area and frequency are minimal.
△ Less
Submitted 20 November, 2019;
originally announced November 2019.
-
The CALOCUBE project for a space based cosmic ray experiment: design, construction, and first performance of a high granularity calorimeter prototype
Authors:
Adriani O.,
Albergo S.,
Auditore L.,
Basti A.,
Berti E.,
Bigongiari G.,
Bonechi L.,
Bongi M.,
Bonvicini V.,
Bottai S.,
Brogi P.,
Cappello G.,
Carotenuto G.,
Castellini G.,
Cattaneo P. W.,
Cecchi R.,
Checchia C.,
D'Alessandro R.,
Detti S.,
Fasoli M.,
Finetti N.,
Italiano A.,
Lenzi P.,
Maestro P.,
Manetti M.
, et al. (26 additional authors not shown)
Abstract:
Current research in High Energy Cosmic Ray Physics touches on fundamental questions regarding the origin of cosmic rays, their composition, the acceleration mechanisms, and their production. Unambiguous measurements of the energy spectra and of the composition of cosmic rays at the "knee" region could provide some of the answers to the above questions. So far only ground based observations, which…
▽ More
Current research in High Energy Cosmic Ray Physics touches on fundamental questions regarding the origin of cosmic rays, their composition, the acceleration mechanisms, and their production. Unambiguous measurements of the energy spectra and of the composition of cosmic rays at the "knee" region could provide some of the answers to the above questions. So far only ground based observations, which rely on sophisticated models describing high energy interactions in the earth's atmosphere, have been possible due to the extremely low particle rates at these energies. A calorimetry based space experiment that could provide not only flux measurements but also energy spectra and particle identification, would certainly overcome some of the uncertainties of ground based experiments. Given the expected particle fluxes, a very large acceptance is needed to collect a sufficient quantity of data, in a time compatible with the duration of a space mission. This in turn, contrasts with the lightness and compactness requirements for space based experiments. We present a novel idea in calorimetry which addresses these issues whilst limiting the mass and volume of the detector. In this paper we report on a four year R&D program where we investigated materials, coatings, photo-sensors, Front End electronics, and mechanical structures with the aim of designing a high performance, high granularity calorimeter with the largest possible acceptance. Details are given of the design choices, component characterisation, and of the construction of a sizeable prototype (Calocube) which has been used in various tests with particle beams.
△ Less
Submitted 22 October, 2019;
originally announced October 2019.
-
Cálculo de constantes ópticas de peliculas delgadas de Silicio compensado a través del método de Swanepoel
Authors:
W. A. Rojas C.,
M. Chaparro P.,
L. M. Pardo C.,
M. I. Cruz F.
Abstract:
Characterization and determination of the optical constants for thin film silicon compensated through the transmittance spectrum is presented. Such properties were determined by R. Swanepoel model. Comparison between the refractive indices of pure silicon and silicon compensated. Increasing refractive index behavior at low wavelengths in both plots was observed. The difference lies in the concentr…
▽ More
Characterization and determination of the optical constants for thin film silicon compensated through the transmittance spectrum is presented. Such properties were determined by R. Swanepoel model. Comparison between the refractive indices of pure silicon and silicon compensated. Increasing refractive index behavior at low wavelengths in both plots was observed. The difference lies in the concentration and type of element that has been added to the thin film. The behavior of the dielectric constant was studied, their relationship with the refractive index. It is found that the value of the dielectric constant to a thin film of silicon compensated agree with those reported in the literature. The reported value of the gap for pure silicon corresponds to $ 1.11eV $ and the value of the gap for the sample corresponds to $ 0.7275eV $, the discrepancy is due to the level of concentration in the compensated silicon film.
△ Less
Submitted 29 July, 2019;
originally announced August 2019.
-
Learning a latent pattern of heterogeneity in the innovation rates of a time series of counts
Authors:
Helton Graziadei,
Hedibert F. Lopes,
Paulo C. Marques F
Abstract:
We develop a Bayesian hierarchical semiparametric model for phenomena related to time series of counts. The main feature of the model is its capability to learn a latent pattern of heterogeneity in the distribution of the process innovation rates, which are softly clustered through time with the help of a Dirichlet process placed at the top of the model hierarchy. The probabilistic forecasting cap…
▽ More
We develop a Bayesian hierarchical semiparametric model for phenomena related to time series of counts. The main feature of the model is its capability to learn a latent pattern of heterogeneity in the distribution of the process innovation rates, which are softly clustered through time with the help of a Dirichlet process placed at the top of the model hierarchy. The probabilistic forecasting capabilities of the model are put to test in the analysis of crime data in Pittsburgh, with favorable results.
△ Less
Submitted 6 July, 2019;
originally announced July 2019.
-
Predictive analysis of microarray data
Authors:
Paulo C. Marques F.,
Carlos A. de B. Pereira
Abstract:
Microarray gene expression data are analyzed by means of a Bayesian nonparametric model, with emphasis on prediction of future observables, yielding a method for selection of differentially expressed genes and a classifier.
Microarray gene expression data are analyzed by means of a Bayesian nonparametric model, with emphasis on prediction of future observables, yielding a method for selection of differentially expressed genes and a classifier.
△ Less
Submitted 10 June, 2014; v1 submitted 8 December, 2013;
originally announced December 2013.
-
On the computation of the marginal likelihood
Authors:
Paulo C. Marques F
Abstract:
We describe briefly in this note a procedure for consistently estimating the marginal likelihood of a statistical model through a sample from the posterior distribution of the model parameters.
We describe briefly in this note a procedure for consistently estimating the marginal likelihood of a statistical model through a sample from the posterior distribution of the model parameters.
△ Less
Submitted 10 June, 2014; v1 submitted 3 June, 2013;
originally announced June 2013.
-
Bayesian Analysis of Simple Random Densities
Authors:
Paulo C. Marques F.,
Carlos A. de B. Pereira
Abstract:
A tractable nonparametric prior over densities is introduced which is closed under sampling and exhibits proper posterior asymptotics.
A tractable nonparametric prior over densities is introduced which is closed under sampling and exhibits proper posterior asymptotics.
△ Less
Submitted 10 June, 2014; v1 submitted 21 September, 2012;
originally announced September 2012.
-
New HST WFC3/UVIS observations augment the stellar-population complexity of omega Centauri
Authors:
Bellini A.,
Bedin L. R.,
Piotto G.,
Milone A. P.,
Marino A. F.,
Villanova S.
Abstract:
We used archival multi-band Hubble Space Telescope observations obtained with the Wide-Field Camera 3 in the UV-optical channel to present new important observational findings on the color-magnitude diagram (CMD) of the Galactic globular cluster omega Centauri. The ultraviolet WFC3 data have been coupled with available WFC/ACS optical-band data. The new CMDs, obtained from the combination of color…
▽ More
We used archival multi-band Hubble Space Telescope observations obtained with the Wide-Field Camera 3 in the UV-optical channel to present new important observational findings on the color-magnitude diagram (CMD) of the Galactic globular cluster omega Centauri. The ultraviolet WFC3 data have been coupled with available WFC/ACS optical-band data. The new CMDs, obtained from the combination of colors coming from eight different bands, disclose an even more complex stellar population than previously identified. This paper discusses the detailed morphology of the CMDs.
△ Less
Submitted 21 June, 2010;
originally announced June 2010.
-
Coupled-channels analyses for large-angle quasi-elastic scattering in massive systems
Authors:
Muhammad Zamrun F.,
K. Hagino,
S. Mitsuoka,
H. Ikezoe
Abstract:
We discuss in detail the coupled-channels approach for the large-angle quasi-elastic scattering in massive systems, where many degrees of freedom may be involved in the reaction. We especially investigate the effects of single, double and triple phonon excitations on the quasi-elastic scattering for $^{48}$Ti,$^{54}$Cr,$^{56}$Fe,$^{64}$Ni and $^{70}$Zn$+^{208}$Pb systems, for which the experimen…
▽ More
We discuss in detail the coupled-channels approach for the large-angle quasi-elastic scattering in massive systems, where many degrees of freedom may be involved in the reaction. We especially investigate the effects of single, double and triple phonon excitations on the quasi-elastic scattering for $^{48}$Ti,$^{54}$Cr,$^{56}$Fe,$^{64}$Ni and $^{70}$Zn$+^{208}$Pb systems, for which the experimental cross sections have been measured recently. We show that the present coupled-channels calculations well account for the overall width of the experimental barrier distribution for these systems. In particular, it is shown that the calculations taking into account single quadrupole phonon excitations in $^{48}$Ti and triple octupole phonon excitations in $^{208}$Pb reasonably well reproduce the experimental quasi-elastic cross section and barrier distribution for the $^{48}$Ti$+^{208}$Pb reaction. On the other hand, $^{54}$Cr,$^{56}$Fe,$^{64}$Ni and $^{70}$Zn$+^{208}$Pb systems seem to require the double quadrupole phonon excitations in the projectiles in order to reproduce the experimental data.
△ Less
Submitted 3 December, 2007;
originally announced December 2007.
-
Effects of anharmonic vibration on large-angle quasi-elastic scattering of 16O+144Sm
Authors:
Muhammad Zamrun F.,
K. Hagino
Abstract:
We study the effects of double octupole and quadrupole phonon excitations in the 144Sm nucleus on quasi-elastic 16O+144Sm scattering at backward angles. To this end, we use the coupled-channels framework, taking into account explicitly the anharmonicities of the vibrations. We use the same coupling scheme as that previously employed to explain the experimental data of sub-barrier fusion cross se…
▽ More
We study the effects of double octupole and quadrupole phonon excitations in the 144Sm nucleus on quasi-elastic 16O+144Sm scattering at backward angles. To this end, we use the coupled-channels framework, taking into account explicitly the anharmonicities of the vibrations. We use the same coupling scheme as that previously employed to explain the experimental data of sub-barrier fusion cross sections for the same system. We show that the experimental data for the quasi-elastic cross sections are well reproduced in this way, although the quasi-elastic barrier distribution has a distinct high energy peak which is somewhat smeared in the experimental barrier distribution. We also discuss the effects of proton transfer on the quasi-elastic barrier distribution. Our study indicates that the fusion and quasi-elastic barrier distributions for this system cannot be accounted for simultaneously with the standard coupled-channels approach.
△ Less
Submitted 15 October, 2007;
originally announced October 2007.
