Showing 1–50 of 81 results for author: Vicente, R

Elastic rods and elastic spinning rings as gravitational wave detectors

Authors: José Natário, Amol Sasane, Rodrigo Vicente

Abstract: Linearised relativistic elasticity equations of motion are considered for a rod and a spinning ring encountering a gravitational wave. In the case of the rod, the equations reduce to a wave equation with appropriate boundary conditions. Using Fourier transforms, the resonant frequencies are found and an explicit distributional solution is given, both for a plus- and a cross-polarised gravitational… ▽ More Linearised relativistic elasticity equations of motion are considered for a rod and a spinning ring encountering a gravitational wave. In the case of the rod, the equations reduce to a wave equation with appropriate boundary conditions. Using Fourier transforms, the resonant frequencies are found and an explicit distributional solution is given, both for a plus- and a cross-polarised gravitational wave. In the case of the spinning ring, the equations are coupled wave equations with periodic boundary conditions. Using a Fourier series expansion, the system of wave equations is recast as a family of ordinary differential equations for the Fourier coefficients, which are then solved via Fourier transforms. The resonant frequencies are found, including simple approximate expressions for slowly rotating rings, and an explicit distributional solution is obtained in the case of the non-spinning ring. Interestingly, it is possible to tune the resonant frequencies by adjusting the angular velocity of the spinning ring. △ Less

Submitted 10 July, 2024; originally announced July 2024.

Comments: 29 pages, 1 figure

MSC Class: Primary 83C10; Secondary 83C35; 83C55; 35A22

Exploring Commonalities in Explanation Frameworks: A Multi-Domain Survey Analysis

Authors: Eduard Barbu, Marharytha Domnich, Raul Vicente, Nikos Sakkas, André Morim

Abstract: This study presents insights gathered from surveys and discussions with specialists in three domains, aiming to find essential elements for a universal explanation framework that could be applied to these and other similar use cases. The insights are incorporated into a software tool that utilizes GP algorithms, known for their interpretability. The applications analyzed include a medical scenario… ▽ More This study presents insights gathered from surveys and discussions with specialists in three domains, aiming to find essential elements for a universal explanation framework that could be applied to these and other similar use cases. The insights are incorporated into a software tool that utilizes GP algorithms, known for their interpretability. The applications analyzed include a medical scenario (involving predictive ML), a retail use case (involving prescriptive ML), and an energy use case (also involving predictive ML). We interviewed professionals from each sector, transcribing their conversations for further analysis. Additionally, experts and non-experts in these fields filled out questionnaires designed to probe various dimensions of explanatory methods. The findings indicate a universal preference for sacrificing a degree of accuracy in favor of greater explainability. Additionally, we highlight the significance of feature importance and counterfactual explanations as critical components of such a framework. Our questionnaires are publicly available to facilitate the dissemination of knowledge in the field of XAI. △ Less

Submitted 20 May, 2024; originally announced May 2024.

COIN: Counterfactual inpainting for weakly supervised semantic segmentation for medical images

Authors: Dmytro Shvetsov, Joonas Ariva, Marharyta Domnich, Raul Vicente, Dmytro Fishman

Abstract: Deep learning is dramatically transforming the field of medical imaging and radiology, enabling the identification of pathologies in medical images, including computed tomography (CT) and X-ray scans. However, the performance of deep learning models, particularly in segmentation tasks, is often limited by the need for extensive annotated datasets. To address this challenge, the capabilities of wea… ▽ More Deep learning is dramatically transforming the field of medical imaging and radiology, enabling the identification of pathologies in medical images, including computed tomography (CT) and X-ray scans. However, the performance of deep learning models, particularly in segmentation tasks, is often limited by the need for extensive annotated datasets. To address this challenge, the capabilities of weakly supervised semantic segmentation are explored through the lens of Explainable AI and the generation of counterfactual explanations. The scope of this research is development of a novel counterfactual inpainting approach (COIN) that flips the predicted classification label from abnormal to normal by using a generative model. For instance, if the classifier deems an input medical image X as abnormal, indicating the presence of a pathology, the generative model aims to inpaint the abnormal region, thus reversing the classifier's original prediction label. The approach enables us to produce precise segmentations for pathologies without depending on pre-existing segmentation masks. Crucially, image-level labels are utilized, which are substantially easier to acquire than creating detailed segmentation masks. The effectiveness of the method is demonstrated by segmenting synthetic targets and actual kidney tumors from CT images acquired from Tartu University Hospital in Estonia. The findings indicate that COIN greatly surpasses established attribution methods, such as RISE, ScoreCAM, and LayerCAM, as well as an alternative counterfactual explanation method introduced by Singla et al. This evidence suggests that COIN is a promising approach for semantic segmentation of tumors in CT images, and presents a step forward in making deep learning applications more accessible and effective in healthcare, where annotated data is scarce. △ Less

Submitted 19 April, 2024; originally announced April 2024.

Comments: This work has been accepted to be presented to The 2nd World Conference on eXplainable Artificial Intelligence (xAI 2024), July 17-19, 2024 - Valletta, Malta

Enhancing Counterfactual Explanation Search with Diffusion Distance and Directional Coherence

Authors: Marharyta Domnich, Raul Vicente

Abstract: A pressing issue in the adoption of AI models is the increasing demand for more human-centric explanations of their predictions. To advance towards more human-centric explanations, understanding how humans produce and select explanations has been beneficial. In this work, inspired by insights of human cognition we propose and test the incorporation of two novel biases to enhance the search for eff… ▽ More A pressing issue in the adoption of AI models is the increasing demand for more human-centric explanations of their predictions. To advance towards more human-centric explanations, understanding how humans produce and select explanations has been beneficial. In this work, inspired by insights of human cognition we propose and test the incorporation of two novel biases to enhance the search for effective counterfactual explanations. Central to our methodology is the application of diffusion distance, which emphasizes data connectivity and actionability in the search for feasible counterfactual explanations. In particular, diffusion distance effectively weights more those points that are more interconnected by numerous short-length paths. This approach brings closely connected points nearer to each other, identifying a feasible path between them. We also introduce a directional coherence term that allows the expression of a preference for the alignment between the joint and marginal directional changes in feature space to reach a counterfactual. This term enables the generation of counterfactual explanations that align with a set of marginal predictions based on expectations of how the outcome of the model varies by changing one feature at a time. We evaluate our method, named Coherent Directional Counterfactual Explainer (CoDiCE), and the impact of the two novel biases against existing methods such as DiCE, FACE, Prototypes, and Growing Spheres. Through a series of ablation experiments on both synthetic and real datasets with continuous and mixed-type features, we demonstrate the effectiveness of our method. △ Less

Submitted 19 April, 2024; originally announced April 2024.

Comments: This work has been accepted to be presented to The 2nd World Conference on eXplainable Artificial Intelligence (xAI 2024), July 17-19, 2024 - Valletta, Malta

Probing the quantum nature of black holes with ultra-light boson environments

Authors: Soumodeep Mitra, Sumanta Chakraborty, Rodrigo Vicente, Justin C. Feng

Abstract: Quantum black holes (BHs), thought of as an excited multilevel system, can be effectively modelled by replacing an event horizon with a partially reflective membrane. This emergent feature affects their interaction with hosting environments, with the most pronounced effects happening for environments of particles with mass~$m_{\rm p}\lesssim (10 M_{\odot}/M)\,10^{-11}\mathrm{\,eV/c^2}$, with~$M$ t… ▽ More Quantum black holes (BHs), thought of as an excited multilevel system, can be effectively modelled by replacing an event horizon with a partially reflective membrane. This emergent feature affects their interaction with hosting environments, with the most pronounced effects happening for environments of particles with mass~$m_{\rm p}\lesssim (10 M_{\odot}/M)\,10^{-11}\mathrm{\,eV/c^2}$, with~$M$ the BH mass. We show that ultra-light bosons -- a viable dark matter candidate -- can be used to probe the quantum nature of BHs. We derive analytical expressions for the accretion rate and dynamical friction acting on exotic compact objects moving through an ultra-light scalar field, which we then use to estimate the orbital de-phasing in the inspiralling of different Laser Inteferometer Space Antenna (LISA) binaries. Our results indicate that LISA may be able to discriminate quantum from classical BHs. △ Less

Submitted 20 May, 2024; v1 submitted 11 December, 2023; originally announced December 2023.

Comments: v2, 15 pages, 2 figures

Axion Weak Leaks: extreme mass-ratio inspirals in ultra-light dark matter

Authors: Francisco Duque, Caio F. B. Macedo, Rodrigo Vicente, Vitor Cardoso

Abstract: Previous works have argued that future gravitational-wave detectors will be able to probe the properties of astrophysical environments where binary coalesce, including accretion disks, but also dark matter structures. Most analyses have resorted to a Newtonian modelling of the environmental effects, which are not suited to study extreme-mass-ratio inspirals immersed in structures of ultra-light bo… ▽ More Previous works have argued that future gravitational-wave detectors will be able to probe the properties of astrophysical environments where binary coalesce, including accretion disks, but also dark matter structures. Most analyses have resorted to a Newtonian modelling of the environmental effects, which are not suited to study extreme-mass-ratio inspirals immersed in structures of ultra-light bosons. In this letter, we use relativistic perturbation theory to consistently study these systems in spherical symmetry. We compute the flux of scalar particles and the rate at which orbital energy (and angular momentum) is dissipated via gravitational radiation and depletion of scalars, i.e. dynamical friction. Our results suggest that the Laser Inteferometer Space Antenna will be able to probe ultra-light dark matter structures in the Galaxy by tracking the phase of extreme-mass-ratio inspirals. △ Less

Submitted 11 December, 2023; originally announced December 2023.

Comments: 5 pages, 2 figures + supplemental material 9 pages, 2 figures

MadQCI: a heterogeneous and scalable SDN QKD network deployed in production facilities

Authors: V. Martin, J. P. Brito, L. Ortiz, R. B. Mendez, J. S. Buruaga, R. J. Vicente, A. Sebastián-Lombraña, D. Rincon, F. Perez, C. Sanchez, M. Peev, H. H. Brunner, F. Fung, A. Poppe, F. Fröwis, A. J. Shields, R. I. Woodward, H. Griesser, S. Roehrich, F. De La Iglesia, C. Abellan, M. Hentschel, J. M. Rivas-Moscoso, A. Pastor, J. Folgueira , et al. (1 additional authors not shown)

Abstract: Current quantum key distribution (QKD) networks focus almost exclusively on transporting secret keys with the highest possible rate. Consequently, they are built as mostly fixed, ad hoc, logically, and physically isolated infrastructures designed to avoid any penalty to the quantum channel. This architecture is neither scalable nor cost-effective and future, real-world deployments will differ cons… ▽ More Current quantum key distribution (QKD) networks focus almost exclusively on transporting secret keys with the highest possible rate. Consequently, they are built as mostly fixed, ad hoc, logically, and physically isolated infrastructures designed to avoid any penalty to the quantum channel. This architecture is neither scalable nor cost-effective and future, real-world deployments will differ considerably. The structure of the MadQCI QKD network presented here is based on disaggregated components and modern paradigms especially designed for flexibility, upgradability, and facilitating the integration of QKD in the security and telecommunications-networks ecosystem. These underlying ideas have been tested by deploying many QKD systems from several manufacturers in a real-world, multi-tenant telecommunications network, installed in production facilities and sharing the infrastructure with commercial traffic. Different technologies have been used in different links to address the variety of situations and needs that arise in real networks, exploring a wide range of possibilities. Finally, a set of realistic use cases have been implemented to demonstrate the validity and performance of the network. The testing took place during a period close to three years, where most of the nodes were continuously active. △ Less

Submitted 3 December, 2023; v1 submitted 21 November, 2023; originally announced November 2023.

Linking QKD testbeds across Europe

Authors: Max Brauer, Rafael J. Vicente, Jaime S. Buruaga, Ruben B. Mendez, Ralf-Peter Braun, Marc Geitz, Piotr Rydlichkowski, Hans H. Brunner, Fred Fung, Momtchil Peev, Antonio Pastor, Diego Lopez, Vicente Martin, Juan P. Brito

Abstract: Quantum-key-distribution (QKD) networks are gaining importance and it has become necessary to analyze the most appropriate methods for their long-distance interconnection. In this paper, four different methods of interconnecting remote QKD networks are proposed. The methods are used to link three different QKD testbeds in Europe, located in Berlin, Madrid, and Poznan. Although long-distance QKD li… ▽ More Quantum-key-distribution (QKD) networks are gaining importance and it has become necessary to analyze the most appropriate methods for their long-distance interconnection. In this paper, four different methods of interconnecting remote QKD networks are proposed. The methods are used to link three different QKD testbeds in Europe, located in Berlin, Madrid, and Poznan. Although long-distance QKD links are only emulated, the used methods can serve as a blueprint for a secure interconnection of distant QKD networks in the future. Specifically, the presented approaches combine, in a transparent way, different fiber and satellite physical media, as well as common standards of key-delivery interfaces. The testbed interconnections are designed to increase the security by utilizing multipath techniques and multiple hybridizations of QKD and post quantum cryptography (PQC) algorithms. △ Less

Submitted 10 January, 2024; v1 submitted 14 November, 2023; originally announced November 2023.

First constraints on compact binary environments from LIGO-Virgo data

Authors: Giada Caneva Santoro, Soumen Roy, Rodrigo Vicente, Maria Haney, Ornella Juliana Piccinni, Walter Del Pozzo, Mario Martinez

Abstract: The LIGO-Virgo analysis of the signals from compact binary mergers observed so far have assumed in-vacuum isolated binary systems, neglecting the potential presence of astrophysical environments. Non-trivial environments may alter gravitational-wave emission, leaving imprints that can be observable via a characteristic dephasing of the emitted signal with respect to the vacuum scenario. We present… ▽ More The LIGO-Virgo analysis of the signals from compact binary mergers observed so far have assumed in-vacuum isolated binary systems, neglecting the potential presence of astrophysical environments. Non-trivial environments may alter gravitational-wave emission, leaving imprints that can be observable via a characteristic dephasing of the emitted signal with respect to the vacuum scenario. We present here the first investigation of environmental effects on the events of the first gravitational-wave catalog (GWTC-1) by LIGO-Virgo. We include the effects of accretion and dynamical friction through a post-Newtonian deformation of the inspiral part of the waveform relative to the vacuum one. We find no evidence for the presence of environmental effects in GWTC-1. Most of the events decisively exclude the scenario of dynamical fragmentation of massive stars as their possible formation channel. Our analysis of GW170817 results in the upper bound on the medium density of $\lesssim 21\: \text{g/cm}^3$. We find that environmental effects can substantially bias the recovered parameters in the vacuum model, even when they are not detectable. Our results forecast that the future 2030s detectors Einstein Telescope and B-DECIGO will be able to probe the environmental effects of accretion disk and superradiant boson clouds on compact binaries. △ Less

Submitted 10 October, 2023; v1 submitted 10 September, 2023; originally announced September 2023.

Comments: 13 pages, 7 figures. Comments are welcomed

Report number: LIGO DCC P2300301

On energy extraction from Q-balls and other fundamental solitons

Authors: Vitor Cardoso, Rodrigo Vicente, Zhen Zhong

Abstract: Energy exchange mechanisms have important applications in particle physics, gravity, fluid mechanics, and practically every field in physics. In this letter we show, both in frequency and time domain, that energy enhancement is possible for waves scattering off fundamental solitons (time-periodic localized structures of bosonic fields), without the need for rotation nor translational motion. We us… ▽ More Energy exchange mechanisms have important applications in particle physics, gravity, fluid mechanics, and practically every field in physics. In this letter we show, both in frequency and time domain, that energy enhancement is possible for waves scattering off fundamental solitons (time-periodic localized structures of bosonic fields), without the need for rotation nor translational motion. We use two-dimensional Q-balls as a testbed, providing the correct criteria for energy amplification, as well as the respective amplification factors, and we discuss possible instability mechanisms. Our results lend support to the qualitative picture drawn in ( arXiv:2212.03269 [gr-qc] ); however we show that this enhancement mechanism is not of superradiant-type, but instead a "blueshift-like" energy exchange between scattering states induced by the background Q-ball, which should occur generically for any time-periodic fundamental soliton. This mechanism does not seem to lead to instabilities. △ Less

Submitted 25 July, 2023; originally announced July 2023.

Comments: 6 pages, 4 figures, accepted for publication in Physical Review Letters

Emergence of Adaptive Circadian Rhythms in Deep Reinforcement Learning

Authors: Aqeel Labash, Florian Fletzer, Daniel Majoral, Raul Vicente

Abstract: Adapting to regularities of the environment is critical for biological organisms to anticipate events and plan. A prominent example is the circadian rhythm corresponding to the internalization by organisms of the $24$-hour period of the Earth's rotation. In this work, we study the emergence of circadian-like rhythms in deep reinforcement learning agents. In particular, we deployed agents in an env… ▽ More Adapting to regularities of the environment is critical for biological organisms to anticipate events and plan. A prominent example is the circadian rhythm corresponding to the internalization by organisms of the $24$-hour period of the Earth's rotation. In this work, we study the emergence of circadian-like rhythms in deep reinforcement learning agents. In particular, we deployed agents in an environment with a reliable periodic variation while solving a foraging task. We systematically characterize the agent's behavior during learning and demonstrate the emergence of a rhythm that is endogenous and entrainable. Interestingly, the internal rhythm adapts to shifts in the phase of the environmental signal without any re-training. Furthermore, we show via bifurcation and phase response curve analyses how artificial neurons develop dynamics to support the internalization of the environmental rhythm. From a dynamical systems view, we demonstrate that the adaptation proceeds by the emergence of a stable periodic orbit in the neuron dynamics with a phase response that allows an optimal phase synchronisation between the agent's dynamics and the environmental rhythm. △ Less

Submitted 22 July, 2023; originally announced July 2023.

Comments: ICML 2023

Relativistic drag forces on black holes from scalar dark matter clouds of all sizes

Authors: Dina Traykova, Rodrigo Vicente, Katy Clough, Thomas Helfer, Emanuele Berti, Pedro G. Ferreira, Lam Hui

Abstract: We use numerical simulations of scalar field dark matter evolving on a moving black hole background to confirm the regime of validity of (semi-)analytic expressions derived from first principles for both dynamical friction and momentum accretion in the relativistic regime. We cover both small and large clouds (relative to the de Broglie wavelength of the scalars), and light and heavy particle mass… ▽ More We use numerical simulations of scalar field dark matter evolving on a moving black hole background to confirm the regime of validity of (semi-)analytic expressions derived from first principles for both dynamical friction and momentum accretion in the relativistic regime. We cover both small and large clouds (relative to the de Broglie wavelength of the scalars), and light and heavy particle masses (relative to the BH size). In the case of a small dark matter cloud, the effect of accretion is a non-negligible contribution to the total force on the black hole, even for small scalar masses. We confirm that this momentum accretion transitions between two regimes (wave- and particle-like) and we identify the mass of the scalar at which the transition between regimes occurs. △ Less

Submitted 19 February, 2024; v1 submitted 17 May, 2023; originally announced May 2023.

Comments: 11 pages, 5 figures. Minor corrections and references added to match published version

Journal ref: Phys.Rev.D 108 (2023) 12, L121502

Teaching quantum communications through a hands-on laboratory

Authors: Alberto Sebastián-Lombraña, Laura Ortiz, Juan P. Brito, Jaime Sáez de Buruaga, Rafael J. Vicente, Ruben B. Mendez, Rafael Artiñano, Vicente Martín

Abstract: Teaching quantum communication is a challenging task when involving different technical and engineering backgrounds. The use of an approach that exploits the knowledge of these profiles, as well as other technological resources available for demonstrations or exercises, enhances this teaching. This paper presents as an example the "Quantum Communications Lab" that took place at the 6th INFIERI Sum… ▽ More Teaching quantum communication is a challenging task when involving different technical and engineering backgrounds. The use of an approach that exploits the knowledge of these profiles, as well as other technological resources available for demonstrations or exercises, enhances this teaching. This paper presents as an example the "Quantum Communications Lab" that took place at the 6th INFIERI Summer School in 2021. In this lab, the access to the Madrid Quantum Communication Infrastructure (MadQCI) was an important resource available. △ Less

Submitted 3 May, 2023; v1 submitted 1 March, 2023; originally announced March 2023.

Comments: Accepted for publication in JINST, 12 pages

Journal ref: Jour.Instrumentation 18 (2023) C03026

Can gravitational-wave memory help constrain binary black-hole parameters? A LISA case study

Authors: Silvia Gasparotto, Rodrigo Vicente, Diego Blas, Alexander C. Jenkins, Enrico Barausse

Abstract: Besides the transient effect, the passage of a gravitational wave also causes a persistent displacement in the relative position of an interferometer's test masses through the \emph{nonlinear memory effect}. This effect is generated by the gravitational backreaction of the waves themselves, and encodes additional information about the source. In this work, we explore the implications of using this… ▽ More Besides the transient effect, the passage of a gravitational wave also causes a persistent displacement in the relative position of an interferometer's test masses through the \emph{nonlinear memory effect}. This effect is generated by the gravitational backreaction of the waves themselves, and encodes additional information about the source. In this work, we explore the implications of using this information for the parameter estimation of massive binary black holes with LISA. Based on a Fisher analysis for nonprecessing black hole binaries, our results show that the memory can help to reduce the degeneracy between the luminosity distance and the inclination for binaries observed only for a short time ($\sim$~few hours) before merger. To assess how many such short signals will be detected, we utilized state-of-the-art predictions for the population of massive black hole binaries and models for the gaps expected in the LISA data. We forecast from tens to few hundreds of binaries with observable memory, but only~$\sim \mathcal{O}(0.1)$ events in 4 years for which the memory helps to reduce the degeneracy between distance and inclination. Based on this, we conclude that the new information from the nonlinear memory, while promising for testing general relativity in the strong field regime, has probably a limited impact on further constraining the uncertainty on massive black hole binary parameters with LISA. △ Less

Submitted 28 June, 2023; v1 submitted 30 January, 2023; originally announced January 2023.

Comments: 16 pages, comments are welcomed

Parasitic black holes: the swallowing of a fuzzy dark matter soliton

Authors: Vitor Cardoso, Taishi Ikeda, Rodrigo Vicente, Miguel Zilhão

Abstract: Fuzzy dark matter is an exciting alternative to the standard cold dark matter paradigm, reproducing its large scale predictions, while solving most of the existing tension with small scale observations. These models postulate that dark matter is constituted by light bosons and predict the condensation of a solitonic core -- also known as boson star, supported by wave pressure -- at the center of h… ▽ More Fuzzy dark matter is an exciting alternative to the standard cold dark matter paradigm, reproducing its large scale predictions, while solving most of the existing tension with small scale observations. These models postulate that dark matter is constituted by light bosons and predict the condensation of a solitonic core -- also known as boson star, supported by wave pressure -- at the center of halos. However, solitons which host a \emph{parasitic} supermassive black hole are doomed to be swallowed by their guest. It is thus crucial to understand in detail the accretion process. In this work, we use numerical relativity to self-consistently solve the problem of accretion of a boson star by a central black hole, in spherical symmetry. We identify three stages in the process, a {\it boson-quake}, a {\it catastrophic stage} and a linear phase, as well as a general accurate expression for the lifetime of a boson star with an endoparasitic black hole. Lifetimes of these objects can be large enough to allow them to survive until the present time. △ Less

Submitted 23 December, 2022; v1 submitted 19 July, 2022; originally announced July 2022.

Comments: v2. Overall improvement, added Supplemental Material. To appear in Physical Review D Letters. Movies are also available on https://centra.tecnico.ulisboa.pt/network/grit/files/movies/

A unified framework for dataset shift diagnostics

Authors: Felipe Maia Polo, Rafael Izbicki, Evanildo Gomes Lacerda Jr, Juan Pablo Ibieta-Jimenez, Renato Vicente

Abstract: Supervised learning techniques typically assume training data originates from the target population. Yet, in reality, dataset shift frequently arises, which, if not adequately taken into account, may decrease the performance of their predictors. In this work, we propose a novel and flexible framework called DetectShift that quantifies and tests for multiple dataset shifts, encompassing shifts in t… ▽ More Supervised learning techniques typically assume training data originates from the target population. Yet, in reality, dataset shift frequently arises, which, if not adequately taken into account, may decrease the performance of their predictors. In this work, we propose a novel and flexible framework called DetectShift that quantifies and tests for multiple dataset shifts, encompassing shifts in the distributions of $(X, Y)$, $X$, $Y$, $X|Y$, and $Y|X$. DetectShift equips practitioners with insights into data shifts, facilitating the adaptation or retraining of predictors using both source and target data. This proves extremely valuable when labeled samples in the target domain are limited. The framework utilizes test statistics with the same nature to quantify the magnitude of the various shifts, making results more interpretable. It is versatile, suitable for regression and classification tasks, and accommodates diverse data forms - tabular, text, or image. Experimental results demonstrate the effectiveness of DetectShift in detecting dataset shifts even in higher dimensions. △ Less

Submitted 12 September, 2023; v1 submitted 17 May, 2022; originally announced May 2022.

Journal ref: Information Sciences (2023): 119612

New Horizons for Fundamental Physics with LISA

Authors: K. G. Arun, Enis Belgacem, Robert Benkel, Laura Bernard, Emanuele Berti, Gianfranco Bertone, Marc Besancon, Diego Blas, Christian G. Böhmer, Richard Brito, Gianluca Calcagni, Alejandro Cardenas-Avendaño, Katy Clough, Marco Crisostomi, Valerio De Luca, Daniela Doneva, Stephanie Escoffier, Jose Maria Ezquiaga, Pedro G. Ferreira, Pierre Fleury, Stefano Foffa, Gabriele Franciolini, Noemi Frusciante, Juan García-Bellido, Carlos Herdeiro , et al. (116 additional authors not shown)

Abstract: The Laser Interferometer Space Antenna (LISA) has the potential to reveal wonders about the fundamental theory of nature at play in the extreme gravity regime, where the gravitational interaction is both strong and dynamical. In this white paper, the Fundamental Physics Working Group of the LISA Consortium summarizes the current topics in fundamental physics where LISA observations of GWs can be e… ▽ More The Laser Interferometer Space Antenna (LISA) has the potential to reveal wonders about the fundamental theory of nature at play in the extreme gravity regime, where the gravitational interaction is both strong and dynamical. In this white paper, the Fundamental Physics Working Group of the LISA Consortium summarizes the current topics in fundamental physics where LISA observations of GWs can be expected to provide key input. We provide the briefest of reviews to then delineate avenues for future research directions and to discuss connections between this working group, other working groups and the consortium work package teams. These connections must be developed for LISA to live up to its science potential in these areas. △ Less

Submitted 3 May, 2022; originally announced May 2022.

Comments: Accepted in: Living Reviews in Relativity

MSC Class: 83CXX

Mind the gap: Challenges of deep learning approaches to Theory of Mind

Authors: Jaan Aru, Aqeel Labash, Oriol Corcoll, Raul Vicente

Abstract: Theory of Mind is an essential ability of humans to infer the mental states of others. Here we provide a coherent summary of the potential, current progress, and problems of deep learning approaches to Theory of Mind. We highlight that many current findings can be explained through shortcuts. These shortcuts arise because the tasks used to investigate Theory of Mind in deep learning systems have b… ▽ More Theory of Mind is an essential ability of humans to infer the mental states of others. Here we provide a coherent summary of the potential, current progress, and problems of deep learning approaches to Theory of Mind. We highlight that many current findings can be explained through shortcuts. These shortcuts arise because the tasks used to investigate Theory of Mind in deep learning systems have been too narrow. Thus, we encourage researchers to investigate Theory of Mind in complex open-ended environments. Furthermore, to inspire future deep learning systems we provide a concise overview of prior work done in humans. We further argue that when studying Theory of Mind with deep learning, the research's main focus and contribution ought to be opening up the network's representations. We recommend researchers use tools from the field of interpretability of AI to study the relationship between different network components and aspects of Theory of Mind. △ Less

Submitted 12 December, 2022; v1 submitted 30 March, 2022; originally announced March 2022.

Journal ref: Artificial Intelligence Review, 1-16 (2023)

Dynamical friction of black holes in ultralight dark matter

Authors: Rodrigo Vicente, Vitor Cardoso

Abstract: In this work we derive simple closed-form expressions for the dynamical friction acting on black holes moving through ultralight (scalar field) dark matter, covering both non-relativistic and relativistic black hole speeds. Our derivation is based on long known scattering amplitudes in black hole spacetimes, it includes the effect of black hole spin and can be easily extended to vector and tensor… ▽ More In this work we derive simple closed-form expressions for the dynamical friction acting on black holes moving through ultralight (scalar field) dark matter, covering both non-relativistic and relativistic black hole speeds. Our derivation is based on long known scattering amplitudes in black hole spacetimes, it includes the effect of black hole spin and can be easily extended to vector and tensor light fields. Our results cover and complement recent numerical and previous non-relativistic treatments of dynamical friction in ultralight dark matter. △ Less

Submitted 11 April, 2022; v1 submitted 21 January, 2022; originally announced January 2022.

Comments: 11 pages, 2 figures. Version to be published in Physical Review D

LegalNLP -- Natural Language Processing methods for the Brazilian Legal Language

Authors: Felipe Maia Polo, Gabriel Caiaffa Floriano Mendonça, Kauê Capellato J. Parreira, Lucka Gianvechio, Peterson Cordeiro, Jonathan Batista Ferreira, Leticia Maria Paz de Lima, Antônio Carlos do Amaral Maia, Renato Vicente

Abstract: We present and make available pre-trained language models (Phraser, Word2Vec, Doc2Vec, FastText, and BERT) for the Brazilian legal language, a Python package with functions to facilitate their use, and a set of demonstrations/tutorials containing some applications involving them. Given that our material is built upon legal texts coming from several Brazilian courts, this initiative is extremely he… ▽ More We present and make available pre-trained language models (Phraser, Word2Vec, Doc2Vec, FastText, and BERT) for the Brazilian legal language, a Python package with functions to facilitate their use, and a set of demonstrations/tutorials containing some applications involving them. Given that our material is built upon legal texts coming from several Brazilian courts, this initiative is extremely helpful for the Brazilian legal field, which lacks other open and specific tools and language models. Our main objective is to catalyze the use of natural language processing tools for legal texts analysis by the Brazilian industry, government, and academia, providing the necessary tools and accessible material. △ Less

Submitted 5 October, 2021; originally announced October 2021.

The Gravity of Classical Fields: And Its Effect on the Dynamics of Gravitational Systems

Authors: Rodrigo Vicente

Abstract: Classical fields are ubiquitous in theoretical physics. They find applications in almost all areas of physics, from condensed matter and particle physics to cosmology and astrophysics. Scalar fields, in particular, can give rise to confined structures, such as boson stars, oscillatons or Q-balls. These objects are interesting hypothetical new "dark matter stars", but also good descriptions of dark… ▽ More Classical fields are ubiquitous in theoretical physics. They find applications in almost all areas of physics, from condensed matter and particle physics to cosmology and astrophysics. Scalar fields, in particular, can give rise to confined structures, such as boson stars, oscillatons or Q-balls. These objects are interesting hypothetical new "dark matter stars", but also good descriptions of dark matter cores when the fields are ultralight. In this thesis, we study the dynamical response of such confined bosonic structures when excited by external matter (stars, planets or black holes) in their vicinities. Such perturbers can either be piercing through the bosonic configuration or undergoing periodic motion at its center (e.g., binaries). Our setup can also efficiently describe the interaction between a moving massive black hole and the surrounding environment. It also depicts dark matter depletion as a reaction to an inspiralling binary within a dark matter core. Our results provide a complete picture of the interaction between black holes or stars and the ultralight dark matter core environment where they may live in. This thesis also deals with several classical field environmental effects on the motion (or, ultimately, the survival) of compact objects, like black holes. △ Less

Submitted 14 October, 2021; originally announced October 2021.

Comments: PhD thesis in Physics. arXiv admin note: text overlap with arXiv:2009.00012

Effects of personality traits in predicting grade retention of Brazilian students

Authors: Carmen Melo Toledo, Guilherme Mendes Bassedon, Jonathan Batista Ferreira, Lucka de Godoy Gianvechio, Carlos Guatimosim, Felipe Maia Polo, Renato Vicente

Abstract: Student's grade retention is a key issue faced by many education systems, especially those in develo** countries. In this paper, we seek to gauge the relevance of students' personality traits in predicting grade retention in Brazil. For that, we used data collected in 2012 and 2017, in the city of Sertaozinho, countryside of the state of Sao Paulo, Brazil. The surveys taken in Sertaozinho includ… ▽ More Student's grade retention is a key issue faced by many education systems, especially those in develo** countries. In this paper, we seek to gauge the relevance of students' personality traits in predicting grade retention in Brazil. For that, we used data collected in 2012 and 2017, in the city of Sertaozinho, countryside of the state of Sao Paulo, Brazil. The surveys taken in Sertaozinho included several socioeconomic questions, standardized tests, and a personality test. Moreover, students were in grades 4, 5, and 6 in 2012. Our approach was based on training machine learning models on the surveys' data to predict grade retention between 2012 and 2017 using information from 2012 or before, and then using some strategies to quantify personality traits' predictive power. We concluded that, besides proving to be fairly better than a random classifier when isolated, personality traits contribute to prediction even when using socioeconomic variables and standardized tests results. △ Less

Submitted 12 July, 2021; originally announced July 2021.

Did I do that? Blame as a means to identify controlled effects in reinforcement learning

Authors: Oriol Corcoll, Youssef Mohamed, Raul Vicente

Abstract: Identifying controllable aspects of the environment has proven to be an extraordinary intrinsic motivator to reinforcement learning agents. Despite repeatedly achieving State-of-the-Art results, this approach has only been studied as a proxy to a reward-based task and has not yet been evaluated on its own. Current methods are based on action-prediction. Humans, on the other hand, assign blame to t… ▽ More Identifying controllable aspects of the environment has proven to be an extraordinary intrinsic motivator to reinforcement learning agents. Despite repeatedly achieving State-of-the-Art results, this approach has only been studied as a proxy to a reward-based task and has not yet been evaluated on its own. Current methods are based on action-prediction. Humans, on the other hand, assign blame to their actions to decide what they controlled. This work proposes Controlled Effect Network (CEN), an unsupervised method based on counterfactual measures of blame to identify effects on the environment controlled by the agent. CEN is evaluated in a wide range of environments showing that it can accurately identify controlled effects. Moreover, we demonstrate CEN's capabilities as intrinsic motivator by integrating it in the state-of-the-art exploration method, achieving substantially better performance than action-prediction models. △ Less

Submitted 17 February, 2022; v1 submitted 1 June, 2021; originally announced June 2021.

Deep Neural Networks using a Single Neuron: Folded-in-Time Architecture using Feedback-Modulated Delay Loops

Authors: Florian Stelzer, André Röhm, Raul Vicente, Ingo Fischer, Serhiy Yanchuk

Abstract: Deep neural networks are among the most widely applied machine learning tools showing outstanding performance in a broad range of tasks. We present a method for folding a deep neural network of arbitrary size into a single neuron with multiple time-delayed feedback loops. This single-neuron deep neural network comprises only a single nonlinearity and appropriately adjusted modulations of the feedb… ▽ More Deep neural networks are among the most widely applied machine learning tools showing outstanding performance in a broad range of tasks. We present a method for folding a deep neural network of arbitrary size into a single neuron with multiple time-delayed feedback loops. This single-neuron deep neural network comprises only a single nonlinearity and appropriately adjusted modulations of the feedback signals. The network states emerge in time as a temporal unfolding of the neuron's dynamics. By adjusting the feedback-modulation within the loops, we adapt the network's connection weights. These connection weights are determined via a back-propagation algorithm, where both the delay-induced and local network connections must be taken into account. Our approach can fully represent standard Deep Neural Networks (DNN), encompasses sparse DNNs, and extends the DNN concept toward dynamical systems implementations. The new method, which we call Folded-in-time DNN (Fit-DNN), exhibits promising performance in a set of benchmark tasks. △ Less

Submitted 6 June, 2021; v1 submitted 19 November, 2020; originally announced November 2020.

An experiment on the mechanisms of racial bias in ML-based credit scoring in Brazil

Authors: Ramon Vilarino, Renato Vicente

Abstract: We dissect an experimental credit scoring model developed with real data and demonstrate - without access to protected attributes - how the use of location information introduces racial bias. We analyze the tree gradient boosting model with the aid of a game-theoretic inspired machine learning explainability technique, counterfactual experiments and Brazilian census data. By exposing algorithmic r… ▽ More We dissect an experimental credit scoring model developed with real data and demonstrate - without access to protected attributes - how the use of location information introduces racial bias. We analyze the tree gradient boosting model with the aid of a game-theoretic inspired machine learning explainability technique, counterfactual experiments and Brazilian census data. By exposing algorithmic racial bias explaining the trained machine learning model inner mechanisms, this experiment comprises an interesting artifact to aid the endeavor of theoretical understanding of the emergence of racial bias in machine learning systems. Without access to individuals' racial categories, we show how classification parity measures using geographically defined groups could carry information about model racial bias. The experiment testifies to the need for methods and language that do not presuppose access to protected attributes when auditing ML models, the importance of considering regional specifics when addressing racial issues, and the central role of census data in the AI research community. To the best of our knowledge, this is the first documented case of algorithmic racial bias in ML-based credit scoring in Brazil, the country with the second largest Black population in the world. △ Less

Submitted 15 December, 2021; v1 submitted 10 November, 2020; originally announced November 2020.

Eccentricity evolution of compact binaries and applications to gravitational-wave physics

Authors: Vitor Cardoso, Caio F. B. Macedo, Rodrigo Vicente

Abstract: Searches for gravitational waves from compact binaries focus mostly on quasi-circular motion, with the rationale that wave emission circularizes the orbit. Here, we study the generality of this result, when astrophysical environments (e.g., accretion disks) or other fundamental interactions are taken into account. We are motivated by possible electromagnetic counterparts to binary black hole coale… ▽ More Searches for gravitational waves from compact binaries focus mostly on quasi-circular motion, with the rationale that wave emission circularizes the orbit. Here, we study the generality of this result, when astrophysical environments (e.g., accretion disks) or other fundamental interactions are taken into account. We are motivated by possible electromagnetic counterparts to binary black hole coalescences and orbits, but also by the possible use of eccentricity as a smoking-gun for new physics. We find that: i) backreaction from radiative mechanisms, including scalars, vectors and gravitational waves circularize the orbital motion. ii) by contrast, environmental effects such as accretion and dynamical friction increase the eccentricity of binaries. Thus, it is the competition between radiative mechanisms and environmental effects that dictates the eccentricity evolution. We study this competition within an adiabatic approach, including gravitational radiation and dynamical friction forces. We show that that there is a critical semi-major axis below which gravitational radiation dominates the motion and the eccentricity of the system decreases. However, the eccentricity inherited from the environment-dominated stage can be substantial, and in particular can affect LISA sources. We provide examples for GW190521-like sources. △ Less

Submitted 11 January, 2021; v1 submitted 28 October, 2020; originally announced October 2020.

Comments: 12 pages, 3 figures. v4: Version to be published in Physical Review D

Journal ref: Phys. Rev. D 103, 023015 (2021)

Partial Information Decomposition of Boolean Functions: a Fourier Analysis perspective

Authors: Abdullah Makkeh, Dirk Oliver Theis, Raul Vicente

Abstract: Partial information decomposition (PID) partitions the information that a set of sources has about a target variable into synergistic, unique, and redundant contributions. This information-theoretic tool has recently attracted attention due to its potential to characterize the information processing in multivariate systems. However, the PID framework still lacks a solid and intuitive interpretatio… ▽ More Partial information decomposition (PID) partitions the information that a set of sources has about a target variable into synergistic, unique, and redundant contributions. This information-theoretic tool has recently attracted attention due to its potential to characterize the information processing in multivariate systems. However, the PID framework still lacks a solid and intuitive interpretation of its information components. In the aim to improve the understanding of PID components, we focus here on Boolean gates, a much-studied type of source-target mechanisms. Boolean gates have been extensively characterised via Fourier analysis which coefficients have been related to interesting properties of the functions defining the gates. In this paper, we establish for Boolean gates mechanisms a relation between their PID components and Fourier coefficients. △ Less

Submitted 14 October, 2020; originally announced October 2020.

Comments: 24 pages; 1 figure

Disentangling causal effects for hierarchical reinforcement learning

Authors: Oriol Corcoll, Raul Vicente

Abstract: Exploration and credit assignment under sparse rewards are still challenging problems. We argue that these challenges arise in part due to the intrinsic rigidity of operating at the level of actions. Actions can precisely define how to perform an activity but are ill-suited to describe what activity to perform. Instead, causal effects are inherently composable and temporally abstract, making them… ▽ More Exploration and credit assignment under sparse rewards are still challenging problems. We argue that these challenges arise in part due to the intrinsic rigidity of operating at the level of actions. Actions can precisely define how to perform an activity but are ill-suited to describe what activity to perform. Instead, causal effects are inherently composable and temporally abstract, making them ideal for descriptive tasks. By leveraging a hierarchy of causal effects, this study aims to expedite the learning of task-specific behavior and aid exploration. Borrowing counterfactual and normality measures from causal literature, we disentangle controllable effects from effects caused by other dynamics of the environment. We propose CEHRL, a hierarchical method that models the distribution of controllable effects using a Variational Autoencoder. This distribution is used by a high-level policy to 1) explore the environment via random effect exploration so that novel effects are continuously discovered and learned, and to 2) learn task-specific behavior by prioritizing the effects that maximize a given reward function. In comparison to exploring with random actions, experimental results show that random effect exploration is a more efficient mechanism and that by assigning credit to few effects rather than many actions, CEHRL learns tasks more rapidly. △ Less

Submitted 21 February, 2022; v1 submitted 3 October, 2020; originally announced October 2020.

Effective Sample Size, Dimensionality, and Generalization in Covariate Shift Adaptation

Authors: Felipe Maia Polo, Renato Vicente

Abstract: In supervised learning, training and test datasets are often sampled from distinct distributions. Domain adaptation techniques are thus required. Covariate shift adaptation yields good generalization performance when domains differ only by the marginal distribution of features. Covariate shift adaptation is usually implemented using importance weighting, which may fail, according to common wisdom,… ▽ More In supervised learning, training and test datasets are often sampled from distinct distributions. Domain adaptation techniques are thus required. Covariate shift adaptation yields good generalization performance when domains differ only by the marginal distribution of features. Covariate shift adaptation is usually implemented using importance weighting, which may fail, according to common wisdom, due to small effective sample sizes (ESS). Previous research argues this scenario is more common in high-dimensional settings. However, how effective sample size, dimensionality, and model performance/generalization are formally related in supervised learning, considering the context of covariate shift adaptation, is still somewhat obscure in the literature. Thus, a main challenge is presenting a unified theory connecting those points. Hence, in this paper, we focus on building a unified view connecting the ESS, data dimensionality, and generalization in the context of covariate shift adaptation. Moreover, we also demonstrate how dimensionality reduction or feature selection can increase the ESS, and argue that our results support dimensionality reduction before covariate shift adaptation as a good practice. △ Less

Submitted 8 January, 2022; v1 submitted 2 October, 2020; originally announced October 2020.

Comments: Neural Comput & Applic (2022)

The response of ultralight dark matter to supermassive black holes and binaries

Authors: Lorenzo Annulli, Vitor Cardoso, Rodrigo Vicente

Abstract: Scalar fields can give rise to confined structures, such as boson stars or Q-balls. These objects are interesting hypothetical new "dark matter stars," but also good descriptions of dark matter haloes when the fields are ultralight. Here, we study the dynamical response of such confined bosonic structures when excited by external matter (stars, planets or black holes) in their vicinities. Such per… ▽ More Scalar fields can give rise to confined structures, such as boson stars or Q-balls. These objects are interesting hypothetical new "dark matter stars," but also good descriptions of dark matter haloes when the fields are ultralight. Here, we study the dynamical response of such confined bosonic structures when excited by external matter (stars, planets or black holes) in their vicinities. Such perturbers can either be plunging through the bosonic configuration or simply act as periodic sources. Our setup can also efficiently describe the interaction between a massive black hole and the surrounding environment, shortly after the massive body has undergone a "kick", due to the collapse of baryonic matter at the galactic center. It also depicts dark matter depletion as a reaction to an inspiralling binary within the halo. We calculate total energy loss, and linear and angular momenta radiated during these processes, and perform the first self-consistent calculation of dynamical friction acting on moving bodies in these backgrounds. We show that the gravitational collapse to a supermassive black hole at the center of a Newtonian boson star (NBS) is accompanied by a small change in the surrounding core. The NBS eventually gets accreted, but only on times larger than a Hubble scale for astrophysical parameters. Stellar or supermassive binaries are able to "stir" and expel scalar from the NBS. For binaries in the LIGO or LISA band, close to coalescence, scalar emission affects the waveform at leading $-6$ PN order with respect to the dominant quadrupolar term; the coefficient is too small to allow detection by next-generation interferometers. Our results provide a complete picture of the interaction between black holes or stars and the ultralight dark matter environment they live in. △ Less

Submitted 31 August, 2020; originally announced September 2020.

Comments: 38 pages, 16 figures, RevTex4. Comments are welcome

Journal ref: Phys. Rev. D 102, 063022 (2020)

A semi-linear wave model for critical collapse

Authors: Isabel Suárez Fernández, Rodrigo Vicente, David Hilditch

Abstract: In spherical symmetry compelling numerical evidence suggests that in general relativity solutions near the threshold of black hole formation exhibit critical behavior. One aspect of this is that threshold solutions themselves are self-similar and are, in a certain sense, unique. To an extent yet to be fully understood, the same phenomena persist beyond spherical symmetry. It is therefore desirable… ▽ More In spherical symmetry compelling numerical evidence suggests that in general relativity solutions near the threshold of black hole formation exhibit critical behavior. One aspect of this is that threshold solutions themselves are self-similar and are, in a certain sense, unique. To an extent yet to be fully understood, the same phenomena persist beyond spherical symmetry. It is therefore desirable to construct models that exhibit such symmetry at the threshold of blow-up. Starting with deformations of the wave equation, we discuss models which have discretely self-similar threshold solutions. We study threshold solutions in the past light cone of the blow-up point. In spherical symmetry there is a sense in which a unique critical solution exists. Spherical numerical evolutions are also presented for more general models, and exhibit similar behavior. Away from spherical symmetry threshold solutions attain more freedom. Different topologies of blow-up are possible, and even locally the critical solution needs reinterpretation as a parameterized family. △ Less

Submitted 27 July, 2020; originally announced July 2020.

Comments: 18 pages, 9 figures

Journal ref: Phys. Rev. D 103, 044016 (2021)

Stirred and shaken: dynamical behavior of boson stars and dark matter cores

Authors: Lorenzo Annulli, Vitor Cardoso, Rodrigo Vicente

Abstract: Bosonic fields can give rise to self-gravitating structures. These are interesting hypothetical new "dark matter stars" and good descriptions of dark matter haloes if the fields are very light. We study the dynamical response of Newtonian boson stars (NBS) when excited by external matter (stars, planets or black holes) in their vicinities. Our setup can describe the interaction between a massive b… ▽ More Bosonic fields can give rise to self-gravitating structures. These are interesting hypothetical new "dark matter stars" and good descriptions of dark matter haloes if the fields are very light. We study the dynamical response of Newtonian boson stars (NBS) when excited by external matter (stars, planets or black holes) in their vicinities. Our setup can describe the interaction between a massive black hole and the surrounding environment, shortly after the massive body has undergone a "kick", due to the collapse of baryonic matter at the galactic center, or dark matter depletion as a reaction to an inspiralling binary. We perform the first self-consistent calculation of dynamical friction acting on moving bodies in these backgrounds. Binaries close to coalescence "stir" the NBS core, and backreaction affects gravitational waveforms at leading $-6PN$ order with respect to the dominant quadrupolar term; the coefficient is too small to allow detection by next-generation interferometers. We also show that the gravitational collapse to a supermassive black hole at the center of a NBS is accompanied by only a small change in the surrounding core. The NBS eventually gets accreted, but for astrophysical parameters this occurs only after several Hubble times. △ Less

Submitted 8 November, 2020; v1 submitted 7 July, 2020; originally announced July 2020.

Comments: 6 pages, 1 figure. v2: clarifications added on framework. Accepted for publication in PLB

Journal ref: Phys.Lett.B 811 (2020) 135944

Restricted Boltzmann Machine Flows and The Critical Temperature of Ising models

Authors: Rodrigo Veiga, Renato Vicente

Abstract: We explore alternative experimental setups for the iterative sampling (flow) from Restricted Boltzmann Machines (RBM) mapped on the temperature space of square lattice Ising models by a neural network thermometer. This framework has been introduced to explore connections between RBM-based deep neural networks and the Renormalization Group (RG). It has been found that, under certain conditions, the… ▽ More We explore alternative experimental setups for the iterative sampling (flow) from Restricted Boltzmann Machines (RBM) mapped on the temperature space of square lattice Ising models by a neural network thermometer. This framework has been introduced to explore connections between RBM-based deep neural networks and the Renormalization Group (RG). It has been found that, under certain conditions, the flow of an RBM trained with Ising spin configurations approaches in the temperature space a value around the critical one: $ k_B T_c / J \approx 2.269$. In this paper we consider datasets with no information about model topology to argue that a neural network thermometer is not an accurate way to detect whether the RBM has learned scale invariance or not. △ Less

Submitted 29 March, 2022; v1 submitted 17 June, 2020; originally announced June 2020.

Comments: 13 figures. Code available at https://github.com/rodsveiga/rbm_flows_ising

Age-structured estimation of COVID-19 ICU demand from low quality data

Authors: Rodrigo Veiga, Rodrigo Murta, Renato Vicente

Abstract: We sample aggravated cases following age-structured probabilities from confirmed cases and use ICU occupation data to find a subnotification factor. A logistic fit is then employed to project the progression of the COVID-19 epidemic with plateau scenarios taken from locations that have reached this stage. Finally, the logistic curve found is corrected by the subnotification factor and sampled to p… ▽ More We sample aggravated cases following age-structured probabilities from confirmed cases and use ICU occupation data to find a subnotification factor. A logistic fit is then employed to project the progression of the COVID-19 epidemic with plateau scenarios taken from locations that have reached this stage. Finally, the logistic curve found is corrected by the subnotification factor and sampled to project the future demand for ICU beds. △ Less

Submitted 15 June, 2020; v1 submitted 8 June, 2020; originally announced June 2020.

Comments: 6 pages, 3 figures, 4 tables. Code available at https://github.com/rodsveiga/ICU_demand

Prospects for Fundamental Physics with LISA

Authors: Enrico Barausse, Emanuele Berti, Thomas Hertog, Scott A. Hughes, Philippe Jetzer, Paolo Pani, Thomas P. Sotiriou, Nicola Tamanini, Helvi Witek, Kent Yagi, Nicolas Yunes, T. Abdelsalhin, A. Achucarro, K. V. Aelst, N. Afshordi, S. Akcay, L. Annulli, K. G. Arun, I. Ayuso, V. Baibhav, T. Baker, H. Bantilan, T. Barreiro, C. Barrera-Hinojosa, N. Bartolo , et al. (296 additional authors not shown)

Abstract: In this paper, which is of programmatic rather than quantitative nature, we aim to further delineate and sharpen the future potential of the LISA mission in the area of fundamental physics. Given the very broad range of topics that might be relevant to LISA, we present here a sample of what we view as particularly promising directions, based in part on the current research interests of the LISA sc… ▽ More In this paper, which is of programmatic rather than quantitative nature, we aim to further delineate and sharpen the future potential of the LISA mission in the area of fundamental physics. Given the very broad range of topics that might be relevant to LISA, we present here a sample of what we view as particularly promising directions, based in part on the current research interests of the LISA scientific community in the area of fundamental physics. We organize these directions through a "science-first" approach that allows us to classify how LISA data can inform theoretical physics in a variety of areas. For each of these theoretical physics classes, we identify the sources that are currently expected to provide the principal contribution to our knowledge, and the areas that need further development. The classification presented here should not be thought of as cast in stone, but rather as a fluid framework that is amenable to change with the flow of new insights in theoretical physics. △ Less

Submitted 27 April, 2020; v1 submitted 27 January, 2020; originally announced January 2020.

Comments: 22 pages, 1 figure, to appear in General Relativity and Gravitation

Journal ref: Gen.Rel.Grav. 52 (2020) 8, 81

The many faces of deep learning

Authors: Raul Vicente

Abstract: Deep learning has sparked a network of mutual interactions between different disciplines and AI. Naturally, each discipline focuses and interprets the workings of deep learning in different ways. This diversity of perspectives on deep learning, from neuroscience to statistical physics, is a rich source of inspiration that fuels novel developments in the theory and applications of machine learning.… ▽ More Deep learning has sparked a network of mutual interactions between different disciplines and AI. Naturally, each discipline focuses and interprets the workings of deep learning in different ways. This diversity of perspectives on deep learning, from neuroscience to statistical physics, is a rich source of inspiration that fuels novel developments in the theory and applications of machine learning. In this perspective, we collect and synthesize different intuitions scattered across several communities as for how deep learning works. In particular, we will briefly discuss the different perspectives that disciplines across mathematics, physics, computation, and neuroscience take on how deep learning does its tricks. Our discussion on each perspective is necessarily shallow due to the multiple views that had to be covered. The deepness in this case should come from putting all these faces of deep learning together in the reader's mind, so that one can look at the same problem from different angles. △ Less

Submitted 25 August, 2019; originally announced August 2019.

Comments: 18 pages

Test fields cannot destroy extremal de Sitter black holes

Authors: Jose Natario, Rodrigo Vicente

Abstract: We determine the timelike Killing vector field that gives the correct definition of energy for test fields propagating in a Kerr-Newman-de Sitter spacetime, and use this result to prove that test fields cannot destroy extremal Kerr-Newman-de Sitter black holes. We determine the timelike Killing vector field that gives the correct definition of energy for test fields propagating in a Kerr-Newman-de Sitter spacetime, and use this result to prove that test fields cannot destroy extremal Kerr-Newman-de Sitter black holes. △ Less

Submitted 26 August, 2019; originally announced August 2019.

Comments: 13 pages, 3 figures

MSC Class: 83C40; 83C57; 83C75

Journal ref: Gen. Rel. Grav. 52 (2020) 5

Direct information transfer rate optimisation for SSVEP-based BCI

Authors: Anti Ingel, Ilya Kuzovkin, Raul Vicente

Abstract: In this work, a classification method for SSVEP-based BCI is proposed. The classification method uses features extracted by traditional SSVEP-based BCI methods and finds optimal discrimination thresholds for each feature to classify the targets. Optimising the thresholds is formalised as a maximisation task of a performance measure of BCIs called information transfer rate (ITR). However, instead o… ▽ More In this work, a classification method for SSVEP-based BCI is proposed. The classification method uses features extracted by traditional SSVEP-based BCI methods and finds optimal discrimination thresholds for each feature to classify the targets. Optimising the thresholds is formalised as a maximisation task of a performance measure of BCIs called information transfer rate (ITR). However, instead of the standard method of calculating ITR, which makes certain assumptions about the data, a more general formula is derived to avoid incorrect ITR calculation when the standard assumptions are not met. This allows the optimal discrimination thresholds to be automatically calculated and thus eliminates the need for manual parameter selection or performing computationally expensive grid searches. The proposed method shows good performance in classifying targets of a BCI, outperforming previously reported results on the same dataset by a factor of 2 in terms of ITR. The highest achieved ITR on the used dataset was 62 bit/min. The proposed method also provides a way to reduce false classifications, which is important in real-world applications. △ Less

Submitted 19 July, 2019; originally announced July 2019.

Journal ref: Journal of neural engineering, 16(1), 016016 (2018)

Perspective Taking in Deep Reinforcement Learning Agents

Authors: Aqeel Labash, Jaan Aru, Tambet Matiisen, Ardi Tampuu, Raul Vicente

Abstract: Perspective taking is the ability to take the point of view of another agent. This skill is not unique to humans as it is also displayed by other animals like chimpanzees. It is an essential ability for social interactions, including efficient cooperation, competition, and communication. Here we present our progress toward building artificial agents with such abilities. We implemented a perspectiv… ▽ More Perspective taking is the ability to take the point of view of another agent. This skill is not unique to humans as it is also displayed by other animals like chimpanzees. It is an essential ability for social interactions, including efficient cooperation, competition, and communication. Here we present our progress toward building artificial agents with such abilities. We implemented a perspective taking task inspired by experiments done with chimpanzees. We show that agents controlled by artificial neural networks can learn via reinforcement learning to pass simple tests that require perspective taking capabilities. We studied whether this ability is more readily learned by agents with information encoded in allocentric or egocentric form for both their visual perception and motor actions. We believe that, in the long run, building better artificial agents with perspective taking ability can help us develop artificial intelligence that is more human-like and easier to communicate with. △ Less

Submitted 15 April, 2020; v1 submitted 3 July, 2019; originally announced July 2019.

Moving black holes: energy extraction, absorption cross-section and the ring of fire

Authors: Vitor Cardoso, Rodrigo Vicente

Abstract: We consider the interaction between a plane wave and a (counter-moving) black hole. We show that energy is transferred from the black hole to the wave, giving rise to a negative absorption cross-section. Moving black holes absorb radiation and deposit energy in external radiation. Due to this effect, a black hole hole of mass M moving at relativistic speeds in a cold medium will appear to be surro… ▽ More We consider the interaction between a plane wave and a (counter-moving) black hole. We show that energy is transferred from the black hole to the wave, giving rise to a negative absorption cross-section. Moving black holes absorb radiation and deposit energy in external radiation. Due to this effect, a black hole hole of mass M moving at relativistic speeds in a cold medium will appear to be surrounded by a bright "ring" of diameter 3\sqrt{3}GM/c^2 and thickness GM/c^2. △ Less

Submitted 2 August, 2019; v1 submitted 24 June, 2019; originally announced June 2019.

Comments: 6 pages, 3 figures. v2: minor edits and corrections. Added comparison to inverse Compton scattering

Journal ref: Phys. Rev. D 100, 084001 (2019)

Dynamical friction in slab geometries and accretion disks

Authors: Rodrigo Vicente, Vitor Cardoso, Miguel Zilhão

Abstract: The evolution of planets, stars and even galaxies is driven, to a large extent, by dynamical friction of gravitational origin. There is now a good understanding of the friction produced by extended media, either collisionless of fluid-like. However, the physics of accretion or protoplanetary disks, for instance, is described by slab-like geometries instead, compact in one spatial direction. Here,… ▽ More The evolution of planets, stars and even galaxies is driven, to a large extent, by dynamical friction of gravitational origin. There is now a good understanding of the friction produced by extended media, either collisionless of fluid-like. However, the physics of accretion or protoplanetary disks, for instance, is described by slab-like geometries instead, compact in one spatial direction. Here, we find, for the first time, the gravitational wake due to a massive perturber moving through a slab-like medium, describing e.g. accretion disks with sharp transitions. We show that dynamical friction in such environments can be substantially reduced relatively to spatially extended profiles. Finally, we provide simple and accurate expressions for the gravitational drag force felt by the perturber, in both the subsonic and supersonic regime. △ Less

Submitted 15 May, 2019; originally announced May 2019.

Comments: 10 pages, 8 figures

MAXENT3D_PID: An Estimator for the Maximum-entropy Trivariate Partial Information Decomposition

Authors: Abdullah Makkeh, Daniel Chicharro, Dirk Oliver Theis, Raul Vicente

Abstract: Chicharro (2017) introduced a procedure to determine multivariate partial information measures within the maximum entropy framework, separating unique, redundant, and synergistic components of information. Makkeh, Theis, and Vicente (2018) formulated the latter trivariate partial information measure as Cone Programming. In this paper, we present MAXENT3D_PID, a production-quality software that com… ▽ More Chicharro (2017) introduced a procedure to determine multivariate partial information measures within the maximum entropy framework, separating unique, redundant, and synergistic components of information. Makkeh, Theis, and Vicente (2018) formulated the latter trivariate partial information measure as Cone Programming. In this paper, we present MAXENT3D_PID, a production-quality software that computes the trivariate partial information measure based on the Cone Programming model. We describe in detail our software, explain how to use it, and perform some experiments reflecting its accuracy in estimating the trivariate partial information decomposition. △ Less

Submitted 10 January, 2019; originally announced January 2019.

Journal ref: Entropy 21 (9), 862, 2019

APES: a Python toolbox for simulating reinforcement learning environments

Authors: Aqeel Labash, Ardi Tampuu, Tambet Matiisen, Jaan Aru, Raul Vicente

Abstract: Assisted by neural networks, reinforcement learning agents have been able to solve increasingly complex tasks over the last years. The simulation environment in which the agents interact is an essential component in any reinforcement learning problem. The environment simulates the dynamics of the agents' world and hence provides feedback to their actions in terms of state observations and external… ▽ More Assisted by neural networks, reinforcement learning agents have been able to solve increasingly complex tasks over the last years. The simulation environment in which the agents interact is an essential component in any reinforcement learning problem. The environment simulates the dynamics of the agents' world and hence provides feedback to their actions in terms of state observations and external rewards. To ease the design and simulation of such environments this work introduces $\texttt{APES}$, a highly customizable and open source package in Python to create 2D grid-world environments for reinforcement learning problems. $\texttt{APES}$ equips agents with algorithms to simulate any field of vision, it allows the creation and positioning of items and rewards according to user-defined rules, and supports the interaction of multiple agents. △ Less

Submitted 31 August, 2018; originally announced August 2018.

Comments: 6 pages, 8 figures

IDTxl: The Information Dynamics Toolkit xl: a Python package for the efficient analysis of multivariate information dynamics in networks

Authors: Patricia Wollstadt, Joseph T. Lizier, Raul Vicente, Conor Finn, Mario Martínez-Zarzuela, Pedro Mediano, Leonardo Novelli, Michael Wibral

Abstract: The Information Dynamics Toolkit xl (IDTxl) is a comprehensive software package for efficient inference of networks and their node dynamics from multivariate time series data using information theory. IDTxl provides functionality to estimate the following measures: 1) For network inference: multivariate transfer entropy (TE)/Granger causality (GC), multivariate mutual information (MI), bivariate… ▽ More The Information Dynamics Toolkit xl (IDTxl) is a comprehensive software package for efficient inference of networks and their node dynamics from multivariate time series data using information theory. IDTxl provides functionality to estimate the following measures: 1) For network inference: multivariate transfer entropy (TE)/Granger causality (GC), multivariate mutual information (MI), bivariate TE/GC, bivariate MI 2) For analysis of node dynamics: active information storage (AIS), partial information decomposition (PID) IDTxl implements estimators for discrete and continuous data with parallel computing engines for both GPU and CPU platforms. Written for Python3.4.3+. △ Less

Submitted 19 February, 2019; v1 submitted 27 July, 2018; originally announced July 2018.

Comments: 4 pages

Journal ref: Journal of Open Source Software, 4(34), 1081

Black holes, gravitational waves and fundamental physics: a roadmap

Authors: Leor Barack, Vitor Cardoso, Samaya Nissanke, Thomas P. Sotiriou, Abbas Askar, Krzysztof Belczynski, Gianfranco Bertone, Edi Bon, Diego Blas, Richard Brito, Tomasz Bulik, Clare Burrage, Christian T. Byrnes, Chiara Caprini, Masha Chernyakova, Piotr Chrusciel, Monica Colpi, Valeria Ferrari, Daniele Gaggero, Jonathan Gair, Juan Garcia-Bellido, S. F. Hassan, Lavinia Heisenberg, Martin Hendry, Ik Siong Heng , et al. (181 additional authors not shown)

Abstract: The grand challenges of contemporary fundamental physics---dark matter, dark energy, vacuum energy, inflation and early universe cosmology, singularities and the hierarchy problem---all involve gravity as a key component. And of all gravitational phenomena, black holes stand out in their elegant simplicity, while harbouring some of the most remarkable predictions of General Relativity: event horiz… ▽ More The grand challenges of contemporary fundamental physics---dark matter, dark energy, vacuum energy, inflation and early universe cosmology, singularities and the hierarchy problem---all involve gravity as a key component. And of all gravitational phenomena, black holes stand out in their elegant simplicity, while harbouring some of the most remarkable predictions of General Relativity: event horizons, singularities and ergoregions. The hitherto invisible landscape of the gravitational Universe is being unveiled before our eyes: the historical direct detection of gravitational waves by the LIGO-Virgo collaboration marks the dawn of a new era of scientific exploration. Gravitational-wave astronomy will allow us to test models of black hole formation, growth and evolution, as well as models of gravitational-wave generation and propagation. It will provide evidence for event horizons and ergoregions, test the theory of General Relativity itself, and may reveal the existence of new fundamental fields. The synthesis of these results has the potential to radically reshape our understanding of the cosmos and of the laws of Nature. The purpose of this work is to present a concise, yet comprehensive overview of the state of the art in the relevant fields of research, summarize important open problems, and lay out a roadmap for future progress. △ Less

Submitted 1 February, 2019; v1 submitted 13 June, 2018; originally announced June 2018.

Comments: White Paper for the COST action "Gravitational Waves, Black Holes, and Fundamental Physics", 272 pages, 12 figures; v4: updated references and author list. Overall improvements and corrections. To appear in Classical and Quantum Gravity

Do deep reinforcement learning agents model intentions?

Authors: Tambet Matiisen, Aqeel Labash, Daniel Majoral, Jaan Aru, Raul Vicente

Abstract: Inferring other agents' mental states such as their knowledge, beliefs and intentions is thought to be essential for effective interactions with other agents. Recently, multiagent systems trained via deep reinforcement learning have been shown to succeed in solving different tasks, but it remains unclear how each agent modeled or represented other agents in their environment. In this work we test… ▽ More Inferring other agents' mental states such as their knowledge, beliefs and intentions is thought to be essential for effective interactions with other agents. Recently, multiagent systems trained via deep reinforcement learning have been shown to succeed in solving different tasks, but it remains unclear how each agent modeled or represented other agents in their environment. In this work we test whether deep reinforcement learning agents explicitly represent other agents' intentions (their specific aims or goals) during a task in which the agents had to coordinate the covering of different spots in a 2D environment. In particular, we tracked over time the performance of a linear decoder trained to predict the final goal of all agents from the hidden state of each agent's neural network controller. We observed that the hidden layers of agents represented explicit information about other agents' goals, i.e. the target landmark they ended up covering. We also performed a series of experiments, in which some agents were replaced by others with fixed goals, to test the level of generalization of the trained agents. We noticed that during the training phase the agents developed a differential preference for each goal, which hindered generalization. To alleviate the above problem, we propose simple changes to the MADDPG training algorithm which leads to better generalization against unseen agents. We believe that training protocols promoting more active intention reading mechanisms, e.g. by preventing simple symmetry-breaking solutions, is a promising direction towards achieving a more robust generalization in different cooperative and competitive tasks. △ Less

Submitted 21 May, 2018; v1 submitted 15 May, 2018; originally announced May 2018.

What deep learning can tell us about higher cognitive functions like mindreading?

Authors: Jaan Aru, Raul Vicente

Abstract: Can deep learning (DL) guide our understanding of computations happening in biological brain? We will first briefly consider how DL has contributed to the research on visual object recognition. In the main part we will assess whether DL could also help us to clarify the computations underlying higher cognitive functions such as Theory of Mind. In addition, we will compare the objectives and learni… ▽ More Can deep learning (DL) guide our understanding of computations happening in biological brain? We will first briefly consider how DL has contributed to the research on visual object recognition. In the main part we will assess whether DL could also help us to clarify the computations underlying higher cognitive functions such as Theory of Mind. In addition, we will compare the objectives and learning signals of brains and machines, leading us to conclude that simply scaling up the current DL algorithms will most likely not lead to human level Theory of Mind. △ Less

Submitted 4 July, 2019; v1 submitted 28 March, 2018; originally announced March 2018.

The Penrose process, superradiance and ergoregion instabilities

Authors: Rodrigo Vicente, Vitor Cardoso, Jorge C. Lopes

Abstract: Superradiant scattering is a radiation enhancement process that takes place in many contexts, and which has recently found exciting applications in astro and particle physics. In the framework of curved spacetime physics, it has been associated with the classical Penrose process for particles. Superradiance is usually also associated with bosonic fields around geometries with ergoregions and horiz… ▽ More Superradiant scattering is a radiation enhancement process that takes place in many contexts, and which has recently found exciting applications in astro and particle physics. In the framework of curved spacetime physics, it has been associated with the classical Penrose process for particles. Superradiance is usually also associated with bosonic fields around geometries with ergoregions and horizons. These notions are in clear tension however: the Penrose process occurs for horizonless geometries, and particles are composed of fermions. Here, we resolve the tension in its different aspects, by showing that (i) superradiance occurs for self-interacting fermions on flat spacetime; (ii) superradiance occurs also for horizonless geometries, where it leads to an ergoregion instability. Ultracompact, horizonless geometries will usually respond with echoes of growing amplitude, until rotational (or electrostatic) energy is extracted from the object; (iii) the Fourier-domain analysis leads to absence of superradiance when horizons are not present. We elucidate why this analysis fails to give meaningful results; (iv) finally, we show that superradiant, ergoregion instabilities have a particle analog of similar growth timescales and which can power the formation of a structure outside a compact, rotating star. △ Less

Submitted 21 March, 2018; originally announced March 2018.

Comments: 20 pages, accepted for publication in PRD

BROJA-2PID: A robust estimator for bivariate partial information decomposition

Authors: Abdullah Makkeh, Dirk Oliver Theis, Raul Vicente

Abstract: Makkeh, Theis, and Vicente found in [8] that Cone Programming model is the most robust to compute the Bertschinger et al. partial information decompostion (BROJA PID) measure [1]. We developed a production-quality robust software that computes the BROJA PID measure based on the Cone Programming model. In this paper, we prove the important property of strong duality for the Cone Program and prove a… ▽ More Makkeh, Theis, and Vicente found in [8] that Cone Programming model is the most robust to compute the Bertschinger et al. partial information decompostion (BROJA PID) measure [1]. We developed a production-quality robust software that computes the BROJA PID measure based on the Cone Programming model. In this paper, we prove the important property of strong duality for the Cone Program and prove an equivalence between the Cone Program and the original Convex problem. Then describe in detail our software and how to use it.\newline\indent △ Less

Submitted 7 February, 2018; originally announced February 2018.

Journal ref: Entropy 20 (4), 271, 2018

Rotating elastic string loops in flat and black hole spacetimes: stability, cosmic censorship and the Penrose process

Authors: Jose Natario, Leonel Queimada, Rodrigo Vicente

Abstract: We rederive the equations of motion for relativistic strings, that is, one-dimensional elastic bodies whose internal energy depends only on their stretching, and use them to study circular string loops rotating in the equatorial plane of flat and black hole spacetimes. We start by obtaining the conditions for equilibrium, and find that: (i) if the string's longitudinal speed of sound does not exce… ▽ More We rederive the equations of motion for relativistic strings, that is, one-dimensional elastic bodies whose internal energy depends only on their stretching, and use them to study circular string loops rotating in the equatorial plane of flat and black hole spacetimes. We start by obtaining the conditions for equilibrium, and find that: (i) if the string's longitudinal speed of sound does not exceed the speed of light then its radius when rotating in Minkowski's spacetime is always larger than its radius when at rest; (ii) in Minkowski's spacetime, equilibria are linearly stable for rotation speeds below a certain threshold, higher than the string's longitudinal speed of sound, and linearly unstable for some rotation speeds above it; (iii) equilibria are always linearly unstable in Schwarzschild's spacetime. Moreover, we study interactions of a rotating string loop with a Kerr black hole, namely in the context of the weak cosmic censorship conjecture and the Penrose process. We find that: (i) elastic string loops that satisfy the null energy condition cannot overspin extremal black holes; (ii) elastic string loops that satisfy the dominant energy condition cannot increase the maximum efficiency of the usual particle Penrose process; (iii) if the dominant energy condition (but not the weak energy condition) is violated then the efficiency can be increased. This last result hints at the interesting possibility that the dominant energy condition may underlie the well known upper bounds for the efficiencies of energy extraction processes (including, for example, superradiance). △ Less

Submitted 11 October, 2023; v1 submitted 14 December, 2017; originally announced December 2017.

Comments: 32 pages, 6 figures; v2: minor improvements; v3: small changes, matches final published version; v4: typos in the references fixed; v5: missing exponent added to Eq. (88)

MSC Class: 74K05; 83C10; 83C55; 83C57; 83C75

Journal ref: Class. Quantum Grav. 35 (2018) 075003