-
Fourier Basis Density Model
Authors:
Alfredo De la Fuente,
Saurabh Singh,
Johannes Ballé
Abstract:
We introduce a lightweight, flexible and end-to-end trainable probability density model parameterized by a constrained Fourier basis. We assess its performance at approximating a range of multi-modal 1D densities, which are generally difficult to fit. In comparison to the deep factorized model introduced in [1], our model achieves a lower cross entropy at a similar computational budget. In additio…
▽ More
We introduce a lightweight, flexible and end-to-end trainable probability density model parameterized by a constrained Fourier basis. We assess its performance at approximating a range of multi-modal 1D densities, which are generally difficult to fit. In comparison to the deep factorized model introduced in [1], our model achieves a lower cross entropy at a similar computational budget. In addition, we also evaluate our method on a toy compression task, demonstrating its utility in learned compression.
△ Less
Submitted 23 February, 2024;
originally announced February 2024.
-
SaLinA: Sequential Learning of Agents
Authors:
Ludovic Denoyer,
Alfredo de la Fuente,
Song Duong,
Jean-Baptiste Gaya,
Pierre-Alexandre Kamienny,
Daniel H. Thompson
Abstract:
SaLinA is a simple library that makes implementing complex sequential learning models easy, including reinforcement learning algorithms. It is built as an extension of PyTorch: algorithms coded with \SALINA{} can be understood in few minutes by PyTorch users and modified easily. Moreover, SaLinA naturally works with multiple CPUs and GPUs at train and test time, thus being a good fit for the large…
▽ More
SaLinA is a simple library that makes implementing complex sequential learning models easy, including reinforcement learning algorithms. It is built as an extension of PyTorch: algorithms coded with \SALINA{} can be understood in few minutes by PyTorch users and modified easily. Moreover, SaLinA naturally works with multiple CPUs and GPUs at train and test time, thus being a good fit for the large-scale training use cases. In comparison to existing RL libraries, SaLinA has a very low adoption cost and capture a large variety of settings (model-based RL, batch RL, hierarchical RL, multi-agent RL, etc.). But SaLinA does not only target RL practitioners, it aims at providing sequential learning capabilities to any deep learning programmer.
△ Less
Submitted 15 October, 2021;
originally announced October 2021.
-
User Subgrou** in Multicast Massive MIMO over Spatially Correlated Rayleigh Fading Channels
Authors:
Alejandro de la Fuente,
Giovanni Interdonato,
Giuseppe Araniti
Abstract:
Massive multiple-input-multiple-output (MaMIMO) multicasting has received significant attention over the last years. MaMIMO is a key enabler of 5G systems to achieve the extremely demanding data rates of upcoming services. Multicast in the physical layer is an efficient way of serving multiple users, simultaneously demanding the same service and sharing radio resources. This work proposes a subgro…
▽ More
Massive multiple-input-multiple-output (MaMIMO) multicasting has received significant attention over the last years. MaMIMO is a key enabler of 5G systems to achieve the extremely demanding data rates of upcoming services. Multicast in the physical layer is an efficient way of serving multiple users, simultaneously demanding the same service and sharing radio resources. This work proposes a subgrou** strategy of multicast users based on their spatial channel characteristics to improve the channel estimation and precoding processes. We employ max-min fairness (MMF) power allocation strategy to maximize the minimum spectral efficiency (SE) of the multicast service. Additionally, we explore the combination of spatial multiplexing with orthogonal (time/frequency) multiple access. By varying the number of antennas at the base station (BS) and users' spatial distribution, we also provide the optimal subgroup configuration that maximizes the spectral efficiency per subgroup. Finally, we show that serving the multicast users into two orthogonal time/frequency intervals offers better performance than only relying on spatial multiplexing.
△ Less
Submitted 19 February, 2021;
originally announced February 2021.
-
Physical Layer Security based Key Management for LoRaWAN
Authors:
Weinand Andreas,
Andreu G. de la Fuente,
Lipps Christoph,
Karrenbauer Michael
Abstract:
Within this the work applicability of Physical LayerSecurity (PHYSEC) based key management within Long RangeWide Area Network (LoRaWAN) is proposed and evaluatedusing an experimental testbed. Since Internet of Things (IoT)technologies have been arising in past years, they have as wellattracted attention for possible cyber attacks. While LoRaWANalready provides many of the features needed in order…
▽ More
Within this the work applicability of Physical LayerSecurity (PHYSEC) based key management within Long RangeWide Area Network (LoRaWAN) is proposed and evaluatedusing an experimental testbed. Since Internet of Things (IoT)technologies have been arising in past years, they have as wellattracted attention for possible cyber attacks. While LoRaWANalready provides many of the features needed in order to ensuresecurity goals such as data confidentiality and integrity, it lacksin measures such as secure key management and distributionschemes. Since conventional solutions are not feasible here, e.g.due to constraints on payload size and power consumption, wepropose the usage of PHYSEC based session key management,which can provide the respective measures in a more lightweightway. The results derived from our testbed show that it can be apromising alternative approach.
△ Less
Submitted 8 January, 2021;
originally announced January 2021.
-
The large charge expansion and AdS/CFT
Authors:
Anton de la Fuente,
Jann Zosso
Abstract:
The scaling dimensions of charged operators in conformal field theory were recently computed in a large charge expansion. We verify this expansion in a dual AdS model. Specifically, we numerically construct solitonic boson star solutions of Einstein-Maxwell-Scalar theory in AdS$_4$ and find that its mass at large charge reproduces the universal form of the lowest operator dimension in the large U(…
▽ More
The scaling dimensions of charged operators in conformal field theory were recently computed in a large charge expansion. We verify this expansion in a dual AdS model. Specifically, we numerically construct solitonic boson star solutions of Einstein-Maxwell-Scalar theory in AdS$_4$ and find that its mass at large charge reproduces the universal form of the lowest operator dimension in the large U(1) charge sector of the dual 2+1 dimensional CFT.
△ Less
Submitted 13 May, 2020;
originally announced May 2020.
-
Bootstrap** mixed correlators in the 2D Ising model
Authors:
Anton de la Fuente
Abstract:
Bootstrap** mixed correlators in three dimensional conformal field theories with a $\mathbb Z_2$ global symmetry has previously led to a closed allowed region in ($Δ_σ$, $Δ_ε$) space surrounding the 3D Ising model. We repeat that analysis in two dimensions. By further assuming a gap in the spin-2 sector, we also find a closed allowed region in ($Δ_σ$, $Δ_ε$) space surrounding the 2D Ising model.
Bootstrap** mixed correlators in three dimensional conformal field theories with a $\mathbb Z_2$ global symmetry has previously led to a closed allowed region in ($Δ_σ$, $Δ_ε$) space surrounding the 3D Ising model. We repeat that analysis in two dimensions. By further assuming a gap in the spin-2 sector, we also find a closed allowed region in ($Δ_σ$, $Δ_ε$) space surrounding the 2D Ising model.
△ Less
Submitted 22 April, 2019;
originally announced April 2019.
-
Rapid seismic domain transfer: Seismic velocity inversion and modeling using deep generative neural networks
Authors:
Lukas Mosser,
Wouter Kimman,
Jesper Dramsch,
Steve Purves,
Alfredo De la Fuente,
Graham Ganssle
Abstract:
Traditional physics-based approaches to infer sub-surface properties such as full-waveform inversion or reflectivity inversion are time-consuming and computationally expensive. We present a deep-learning technique that eliminates the need for these computationally complex methods by posing the problem as one of domain transfer. Our solution is based on a deep convolutional generative adversarial n…
▽ More
Traditional physics-based approaches to infer sub-surface properties such as full-waveform inversion or reflectivity inversion are time-consuming and computationally expensive. We present a deep-learning technique that eliminates the need for these computationally complex methods by posing the problem as one of domain transfer. Our solution is based on a deep convolutional generative adversarial network and dramatically reduces computation time. Training based on two different types of synthetic data produced a neural network that generates realistic velocity models when applied to a real dataset. The system's ability to generalize means it is robust against the inherent occurrence of velocity errors and artifacts in both training and test datasets.
△ Less
Submitted 22 May, 2018;
originally announced May 2018.
-
The large charge expansion at large N
Authors:
Anton de la Fuente
Abstract:
The scaling dimensions of charged operators in conformal field theory have recently been predicted to exhibit universal behavior in the large charge limit. We verify this behavior in the 2+1 dimensional CPN model. Specifically, we numerically compute the scaling dimensions of the lowest dimension monopole operators with charges Q = 1, 2, ... , 100 to subleading order in large N. The coefficients o…
▽ More
The scaling dimensions of charged operators in conformal field theory have recently been predicted to exhibit universal behavior in the large charge limit. We verify this behavior in the 2+1 dimensional CPN model. Specifically, we numerically compute the scaling dimensions of the lowest dimension monopole operators with charges Q = 1, 2, ... , 100 to subleading order in large N. The coefficients of the large Q expansion are extracted through a fit, and the predicted universal O(1) contribution is verified to the subpercent level.
△ Less
Submitted 1 May, 2018;
originally announced May 2018.
-
Rotating superfluids and spinning charged operators in conformal field theory
Authors:
Gabriel Cuomo,
Anton de la Fuente,
Alexander Monin,
David Pirtskhalava,
Riccardo Rattazzi
Abstract:
We calculate the scaling dimensions of operators with large global charge and spin in 2+1 dimensional conformal field theories. By the state-operator correspondence, these operators correspond to superfluids with vortices and can be systematically studied using effective field theory. As the spin increases from zero to the unitarity bound, the superfluid state corresponding to the lowest dimension…
▽ More
We calculate the scaling dimensions of operators with large global charge and spin in 2+1 dimensional conformal field theories. By the state-operator correspondence, these operators correspond to superfluids with vortices and can be systematically studied using effective field theory. As the spin increases from zero to the unitarity bound, the superfluid state corresponding to the lowest dimension operator passes through three distinct regimes: (1) a single phonon, (2) two vortices, and (3) multiple vortices. We also calculate correlation functions with two such operators and the Noether current.
△ Less
Submitted 6 November, 2017;
originally announced November 2017.
-
4D Scattering Amplitudes and Asymptotic Symmetries from 2D CFT
Authors:
Clifford Cheung,
Anton de la Fuente,
Raman Sundrum
Abstract:
We reformulate the scattering amplitudes of 4D flat space gauge theory and gravity in the language of a 2D CFT on the celestial sphere. The resulting CFT structure exhibits an OPE constructed from 4D collinear singularities, as well as infinite-dimensional Kac-Moody and Virasoro algebras encoding the asymptotic symmetries of 4D flat space. We derive these results by recasting 4D dynamics in terms…
▽ More
We reformulate the scattering amplitudes of 4D flat space gauge theory and gravity in the language of a 2D CFT on the celestial sphere. The resulting CFT structure exhibits an OPE constructed from 4D collinear singularities, as well as infinite-dimensional Kac-Moody and Virasoro algebras encoding the asymptotic symmetries of 4D flat space. We derive these results by recasting 4D dynamics in terms of a convenient foliation of flat space into 3D Euclidean AdS and Lorentzian dS geometries. Tree-level scattering amplitudes take the form of Witten diagrams for a continuum of (A)dS modes, which are in turn equivalent to CFT correlators via the (A)dS/CFT dictionary. The Ward identities for the 2D conserved currents are dual to 4D soft theorems, while the bulk-boundary propagators of massless (A)dS modes are superpositions of the leading and subleading Weinberg soft factors of gauge theory and gravity. In general, the massless (A)dS modes are 3D Chern-Simons gauge fields describing the soft, single helicity sectors of 4D gauge theory and gravity. Consistent with the topological nature of Chern-Simons theory, Aharonov-Bohm effects record the "tracks" of hard particles in the soft radiation, leading to a simple characterization of gauge and gravitational memories. Soft particle exchanges between hard processes define the Kac-Moody level and Virasoro central charge, which are thereby related to the 4D gauge coupling and gravitational strength in units of an infrared cutoff. Finally, we discuss a toy model for black hole horizons via a restriction to the Rindler region.
△ Less
Submitted 23 January, 2017; v1 submitted 2 September, 2016;
originally announced September 2016.
-
Natural Inflation and Quantum Gravity
Authors:
Anton de la Fuente,
Prashant Saraswat,
Raman Sundrum
Abstract:
Cosmic Inflation provides an attractive framework for understanding the early universe and the cosmic microwave background. It can readily involve energies close to the scale at which Quantum Gravity effects become important. General considerations of black hole quantum mechanics suggest nontrivial constraints on any effective field theory model of inflation that emerges as a low-energy limit of q…
▽ More
Cosmic Inflation provides an attractive framework for understanding the early universe and the cosmic microwave background. It can readily involve energies close to the scale at which Quantum Gravity effects become important. General considerations of black hole quantum mechanics suggest nontrivial constraints on any effective field theory model of inflation that emerges as a low-energy limit of quantum gravity, in particular the constraint of the Weak Gravity Conjecture. We show that higher-dimensional gauge and gravitational dynamics can elegantly satisfy these constraints and lead to a viable, theoretically-controlled and predictive class of Natural Inflation models.
△ Less
Submitted 29 January, 2015; v1 submitted 10 December, 2014;
originally announced December 2014.
-
Holography of the BTZ Black Hole, Inside and Out
Authors:
Anton de la Fuente,
Raman Sundrum
Abstract:
We propose a 1+1 dimensional CFT dual structure for quantum gravity and matter on the extended 2+1 dimensional BTZ black hole, realized as a quotient of the Poincare patch of AdS$_3$. The quotient spacetime includes regions beyond the singularity, "whiskers", containing timelike and lightlike closed curves, which at first sight seem unphysical. The spacetime includes the usual AdS-asymptotic bound…
▽ More
We propose a 1+1 dimensional CFT dual structure for quantum gravity and matter on the extended 2+1 dimensional BTZ black hole, realized as a quotient of the Poincare patch of AdS$_3$. The quotient spacetime includes regions beyond the singularity, "whiskers", containing timelike and lightlike closed curves, which at first sight seem unphysical. The spacetime includes the usual AdS-asymptotic boundaries outside the horizons as well as boundary components inside the whiskers. We show that local boundary correlators with some endpoints in the whisker regions: (i) are a protected class of amplitudes, dominated by effective field theory even when the associated Witten diagrams appear to traverse the singularity, (ii) describe well-defined diffeomorphism-invariant quantum gravity amplitudes in BTZ, (iii) sharply probe some of the physics inside the horizon but outside the singularity, and (iv) are equivalent to correlators of specific non-local CFT operators in the standard thermofield entangled state of two CFTs. In this sense, the whisker regions can be considered as purely auxiliary spacetimes in which these useful non-local CFT correlators can be rendered as local boundary correlators, and their diagnostic value more readily understood. Our results follow by first performing a novel reanalysis of the Rindler view of standard AdS/CFT duality on the Poincare patch of AdS, followed by exploiting the simple quotient structure of BTZ which turns the Rindler horizon into the BTZ black hole horizon. While most of our checks are within gravitational effective field theory, we arrive at a fully non-perturbative CFT proposal to probe the UV-sensitive approach to the singularity.
△ Less
Submitted 6 July, 2014; v1 submitted 29 July, 2013;
originally announced July 2013.
-
Diffusive Transport Enhanced by Thermal Velocity Fluctuations
Authors:
A. Donev,
A. de la Fuente,
J. B. Bell,
A. L. Garcia
Abstract:
We study the contribution of advection by thermal velocity fluctuations to the effective diffusion coefficient in a mixture of two indistinguishable fluids. The enhancement of the diffusive transport depends on the system size L and grows as \ln(L/L_0) in quasi two-dimensional systems, while in three dimensions it scales as L_0^{-1}-L^{-1}, where L_0 is a reference length. The predictions of a sim…
▽ More
We study the contribution of advection by thermal velocity fluctuations to the effective diffusion coefficient in a mixture of two indistinguishable fluids. The enhancement of the diffusive transport depends on the system size L and grows as \ln(L/L_0) in quasi two-dimensional systems, while in three dimensions it scales as L_0^{-1}-L^{-1}, where L_0 is a reference length. The predictions of a simple fluctuating hydrodynamics theory are compared to results from particle simulations and a finite-volume solver and excellent agreement is observed. Our results conclusively demonstrate that the nonlinear advective terms need to be retained in the equations of fluctuating hydrodynamics when modeling transport in small-scale finite systems.
△ Less
Submitted 25 April, 2011; v1 submitted 28 March, 2011;
originally announced March 2011.
-
Enhancement of Diffusive Transport by Nonequilibrium Thermal Fluctuations
Authors:
A. Donev,
A. de la Fuente,
J. B. Bell,
A. L. Garcia
Abstract:
We study the contribution of advection by thermal velocity fluctuations to the effective diffusion coefficient in a mixture of two identical fluids. The steady-state diffusive flux in a finite system subject to a concentration gradient is enhanced because of long-range correlations between concentration fluctuations and fluctuations of the velocity parallel to the concentration gradient. The enhan…
▽ More
We study the contribution of advection by thermal velocity fluctuations to the effective diffusion coefficient in a mixture of two identical fluids. The steady-state diffusive flux in a finite system subject to a concentration gradient is enhanced because of long-range correlations between concentration fluctuations and fluctuations of the velocity parallel to the concentration gradient. The enhancement of the diffusive transport depends on the system size L and grows as \ln(L/L_{0}) in quasi-two dimensional systems, while in three dimensions it grows as L_{0}^{-1}-L^{-1}, where L_{0} is a reference length. The predictions of a simple fluctuating hydrodynamics theory, closely related to second-order mode-mode coupling analysis, are compared to results from particle simulations and a finite-volume solver and excellent agreement is observed. We elucidate the direct connection to the long-time tail of the velocity autocorrelation function in finite systems, as well as finite-size corrections employed in molecular dynamics calculations. Our results conclusively demonstrate that the nonlinear advective terms need to be retained in the equations of fluctuating hydrodynamics when modeling transport in small-scale finite systems.
△ Less
Submitted 10 May, 2011; v1 submitted 27 March, 2011;
originally announced March 2011.
-
Arecibo HI Absorption Measurements of Pulsars and the Electron Density at Intermediate Longitudes in the First Galactic Quadrant
Authors:
J. M. Weisberg,
S. Stanimirovic,
K. Xilouris,
A. Hedden,
A. de la Fuente,
S. B. Anderson,
F. A. Jenet
Abstract:
We have used the Arecibo telescope to measure the HI absorption spectra of eight pulsars. We show how kinematic distance measurements depend upon the values of the galactic constants R_o and Theta_o, and we select our preferred current values from the literature. We then derive kinematic distances for the low-latitude pulsars in our sample and electron densities along their lines of sight. We co…
▽ More
We have used the Arecibo telescope to measure the HI absorption spectra of eight pulsars. We show how kinematic distance measurements depend upon the values of the galactic constants R_o and Theta_o, and we select our preferred current values from the literature. We then derive kinematic distances for the low-latitude pulsars in our sample and electron densities along their lines of sight. We combine these measurements with all others in the inner galactic plane visible from Arecibo to study the electron density in this region. The electron density in the interarm range 48 degrees < l < 70 degrees is [0.017 (-0.007,+0.012) (68% c.l.)] cm^(-3). This is 0.75 (-0.22,+0.49) (68% c.l.) of the value calculated by the Cordes & Lazio (2002) galactic electron density model. The model agrees more closely with electron density measurements toward Arecibo pulsars lying closer to the galactic center, at 30 degrees<l<48 degrees. Our analysis leads to the best current estimate of the distance of the relativistic binary pulsar B1913+16: d=(9.0 +/- 3) kpc.
We use the high-latitude pulsars to search for small-scale structure in the interstellar hydrogen observed in absorption over multiple epochs. PSR B0301+19 exhibited significant changes in its absorption spectrum over 22 yr, indicating HI structure on a ~500 AU scale.
△ Less
Submitted 24 September, 2007;
originally announced September 2007.
-
PSR B1849+00 probes the tiny-scale molecular gas?
Authors:
S. Stanimirovic,
J. M. Weisberg,
J. M. Dickey,
A. de la Fuente,
K. Devine,
A. Hedden,
S. B. Anderson
Abstract:
In this paper we present and discuss the great difference in OH absorption spectra against PSR B1849+00 and SNR G33.6+0.1 along the same line-of-sight. This finding is important as it clearly demonstrates that statistics of absorbing molecular gas depends on the size of the background source.
In this paper we present and discuss the great difference in OH absorption spectra against PSR B1849+00 and SNR G33.6+0.1 along the same line-of-sight. This finding is important as it clearly demonstrates that statistics of absorbing molecular gas depends on the size of the background source.
△ Less
Submitted 30 June, 2003;
originally announced July 2003.
-
Detection of OH absorption against PSR B1849+00
Authors:
Snezana Stanimirovic,
Joel M. Weisberg,
John M. Dickey,
Anton de la Fuente,
Kathryn Devine,
Abigail Hedden,
Stuart B. Anderson
Abstract:
We have searched for OH absorption against seven pulsars using the Arecibo telescope. In both OH mainlines (at 1665 and 1667 MHz), deep and narrow absorption features were detected toward PSR B1849+00. In addition, we have detected several absorption and emission features against B33.6+0.1, a nearby supernova remnant (SNR). The most interesting result of this study is that a pencil-sharp absorpt…
▽ More
We have searched for OH absorption against seven pulsars using the Arecibo telescope. In both OH mainlines (at 1665 and 1667 MHz), deep and narrow absorption features were detected toward PSR B1849+00. In addition, we have detected several absorption and emission features against B33.6+0.1, a nearby supernova remnant (SNR). The most interesting result of this study is that a pencil-sharp absorption sample against the PSR differs greatly from the large-angle absorption sample observed against the SNR. If both the PSR and the SNR probe the same molecular cloud then this finding has important implications for absorption studies of the molecular medium, as it shows that the statistics of absorbing OH depends on the size of the background source. We also show that the OH absorption against the PSR most likely originates from a small (<30 arcsec) and dense (>10^5 cm^-3) molecular clump.
△ Less
Submitted 9 April, 2003;
originally announced April 2003.
