Insert related papers here

Insert related papers here

Insert related papers here

Insert related papers here

Insert related papers here

Insert related papers here

Welcome to Byte Size Arxiv

Papers made digestable

2022-11-30

A numerical study on the role of instabilities on multi-wavelength emission signatures of blazar jets

  • In this regard, we have performed RMHD simulations of a representative section of blazar jet
  • We have further compared the effects of different radiation mechanisms through numerical simulation of 2D slab jet as a validation test
  • We observe that shocks produced with the evolution of instabilities give rise to flaring signatures in the high energy band
  • The impact of such shocks is also evident from the instantaneous flattening of the synchrotron component of the SEDs
  • At later stages, we observe the transition in X-ray emission from the synchrotron process to that dominated by EC.
Blazars, a class of active galaxies whose jets are relativistic and collimated flows of plasma directed along the line of sight and are prone to a slew of magneto-hydrodynamic (MHD) instabilities. We aim to study the interplay of radiation and particle acceleration processes in regulating the multi-band emission and variability signatures from blazars. In particular, the goal is to decipher the impact of shocks arising due to MHD instabilities in driving the longterm variable emission signatures from blazars. In this regard, we have performed RMHD simulations of a representative section of blazar jet. The jet is evolved using a hybrid Eulerian-Lagrangian framework to account for radiative losses due to synchrotron process and particle acceleration due to shocks. Additionally, we have incorporated and validated radiative losses due to the external Compton (EC) process that are relevant for blazars. We have further compared the effects of different radiation mechanisms through numerical simulation of 2D slab jet as a validation test. Finally, we have carried out a parametric study to quantify the effect of magnetic fields and external radiation field characteristics by performing 3D simulations of a plasma column. The synthetic light curves and spectral energy distribution (SEDs) are analysed to qualitatively understand the impact of instability driven shocks. We observe that shocks produced with the evolution of instabilities give rise to flaring signatures in the high energy band. The impact of such shocks is also evident from the instantaneous flattening of the synchrotron component of the SEDs. At later stages, we observe the transition in X-ray emission from the synchrotron process to that dominated by EC. The inclusion of the EC process also gives rise to gamma-ray emission and shows signatures of mild Compton dominance as typically seen in Low Synchrotron Peaked blazars.

Authors: Sriyasriti Acharya, Bhargav Vaidya, Indu Kalpa Dihingia, Sushmita Agarwal, Amit Shukla.

2022-11-30

Positive feedback, quenching and sequential super star cluster (SSC) formation in NGC 4945

  • We estimate proto-SSC ages of 5-9.7$\times$10$^4$ yr
  • Heating by the HII regions in the SSC ZAMS phase seems to be rather local
  • The proto-SSCs in NGC 4945 seem to be more evolved than in the starburst galaxy NGC 253
  • We propose that sequential SSC formation can explain the spatial distribution and different ages of the SSCs in both galaxies.
We have used ALMA imaging (resolutions 0.1\arcsec-0.4\arcsec) of ground and vibrationally excited lines of HCN and HC$_3$N toward the nucleus of NGC 4945 to trace the protostellar phase in Super Star Clusters (proto-SSC). Out of the 14 identified SSCs, we find that 8 are in the proto-SSC phase showing vibrational HCN emission with 5 of them also showing vibrational HC$_3$N emission. We estimate proto-SSC ages of 5-9.7$\times$10$^4$ yr. The more evolved ones, with only HCN emission, are close to reach the Zero Age Main Sequence (ZAMS; ages $\gtrsim$10$^5$ yr). The excitation of the parental cloud seems to be related to the SSC evolutionary stage, with high ($\sim$65 K) and low ($\sim$25 K) rotational temperatures for the youngest proto and ZAMS SSCs, respectively. Heating by the HII regions in the SSC ZAMS phase seems to be rather local. The youngest proto-SSCs are located at the edges of the molecular outflow, indicating SSC formation by positive feedback in the shocked regions. The proto-SSCs in NGC 4945 seem to be more evolved than in the starburst galaxy NGC 253. We propose that sequential SSC formation can explain the spatial distribution and different ages of the SSCs in both galaxies.

Authors: Enrica Bellocchi, Jesús Martín-Pintado, Fernando Rico-Villas, Sergio Martín, Izaskun Jiménez-Sierra.

2022-11-30

Variational Laplace Autoencoders

  • We present a novel approach that addresses both challenges
  • First, we focus on ReLU networks with Gaussian output and illustrate their connection to probabilistic PCA.
Variational autoencoders employ an amortized inference model to approximate the posterior of latent variables. However, such amortized variational inference faces two challenges: (1) the limited posterior expressiveness of fully-factorized Gaussian assumption and (2) the amortization error of the inference model. We present a novel approach that addresses both challenges. First, we focus on ReLU networks with Gaussian output and illustrate their connection to probabilistic PCA. Building on this observation, we derive an iterative algorithm that finds the mode of the posterior and apply full-covariance Gaussian posterior approximation centered on the mode. Subsequently, we present a general framework named Variational Laplace Autoencoders (VLAEs) for training deep generative models. Based on the Laplace approximation of the latent variable posterior, VLAEs enhance the expressiveness of the posterior while reducing the amortization error. Empirical results on MNIST, Omniglot, Fashion-MNIST, SVHN and CIFAR10 show that the proposed approach significantly outperforms other recent amortized or iterative methods on the ReLU networks.

Authors: Yookoon Park, Chris Dongjoo Kim, Gunhee Kim.

2022-11-30

Improved spectral projection estimates

  • We more directly make use of pointwise estimates that are implicit in the work of Berard [2] and avoid the use of weak-type spaces that were used in the previous works [6] and [22]
  • This allows us to prove new results for manifolds of negative curvature and some new sharp estimates for tori
  • We also have new and improved techniques in two dimensions for general manifolds of non-positive curvature.
We obtain new improved spectral projection estimates on manifolds of non-positive curvature, including sharp ones for relatively large spectral windows for general tori. Our results are stronger than those in an earlier work of the first and third authors [6], and the arguments have been greatly simplified. We more directly make use of pointwise estimates that are implicit in the work of Berard [2] and avoid the use of weak-type spaces that were used in the previous works [6] and [22]. We also simplify and strengthen the bilinear arguments by exploiting the use of microlocal $L^2\to L^{q_c}$ Kakeya-Nikodym estimates and avoiding the of $L^2\to L^2$ ones as in earlier results. This allows us to prove new results for manifolds of negative curvature and some new sharp estimates for tori. We also have new and improved techniques in two dimensions for general manifolds of non-positive curvature.

Authors: Matthew D. Blair, Xiaoqi Huang, Christopher D. Sogge.

2022-11-30

Interpretability with full complexity by constraining feature information

  • Interpretability is a pressing issue for machine learning
  • The optimal compression of each feature -- at every stage of approximation -- allows fine-grained inspection of the distinctions among feature values that are most impactful for prediction
  • We develop a framework for extracting insight from the spectrum of approximate models and demonstrate its utility on a range of tabular datasets.
Interpretability is a pressing issue for machine learning. Common approaches to interpretable machine learning constrain interactions between features of the input, rendering the effects of those features on a model's output comprehensible but at the expense of model complexity. We approach interpretability from a new angle: constrain the information about the features without restricting the complexity of the model. Borrowing from information theory, we use the Distributed Information Bottleneck to find optimal compressions of each feature that maximally preserve information about the output. The learned information allocation, by feature and by feature value, provides rich opportunities for interpretation, particularly in problems with many features and complex feature interactions. The central object of analysis is not a single trained model, but rather a spectrum of models serving as approximations that leverage variable amounts of information about the inputs. Information is allocated to features by their relevance to the output, thereby solving the problem of feature selection by constructing a learned continuum of feature inclusion-to-exclusion. The optimal compression of each feature -- at every stage of approximation -- allows fine-grained inspection of the distinctions among feature values that are most impactful for prediction. We develop a framework for extracting insight from the spectrum of approximate models and demonstrate its utility on a range of tabular datasets.

Authors: Kieran A. Murphy, Dani S. Bassett.

2022-11-30

Corner Cases of the Generalized Tau Method

  • We can, therefore, provide one-to-one comparisons to traditional collocation and Galerkin methods within the tau framework
  • We then impose additional mutual compatibility conditions to ensure boundary conditions match at shared subsurfaces
  • The tau corrections and compatibility conditions can be fully isotropic and easily incorporated into existing solvers
  • We present the method explicitly for the Poisson equation in two and three dimensions and describe its extension to arbitrary elliptic equations (e.g
  • biharmonic) in any dimension.
Polynomial spectral methods provide fast, accurate, and flexible solvers for broad ranges of PDEs with one bounded dimension, where the incorporation of general boundary conditions is well understood. However, automating extensions to domains with multiple bounded dimensions is challenging because of difficulties in implementing boundary conditions and imposing compatibility conditions at shared edges and corners. Past work has included various workarounds, such as the anisotropic inclusion of partial boundary data at shared edges or approaches that only work for specific boundary conditions. Here we present a general system for imposing boundary and compatibility conditions for elliptic equations on hypercubes. We take an approach based on the generalized tau method, which allows for a wide range of boundary conditions for many types of spectral methods. The generalized tau method has the distinct advantage that the specified polynomial residual determines the exact algebraic solution; afterwards, any stable numerical scheme will find the same result. We can, therefore, provide one-to-one comparisons to traditional collocation and Galerkin methods within the tau framework. As an essential requirement, we add specific tau corrections to the boundary conditions in addition to the bulk PDE. We then impose additional mutual compatibility conditions to ensure boundary conditions match at shared subsurfaces. Our approach works with general boundary conditions that commute on intersecting subsurfaces, including Dirichlet, Neumann, Robin, and any combination of these on all boundaries. The tau corrections and compatibility conditions can be fully isotropic and easily incorporated into existing solvers. We present the method explicitly for the Poisson equation in two and three dimensions and describe its extension to arbitrary elliptic equations (e.g. biharmonic) in any dimension.

Authors: Keaton J. Burns, Daniel Fortunato, Keith Julien, Geoffrey M. Vasil.

2022-11-30

Twice-Marked Banana Graphs & Brill-Noether Generality

  • We analyze a family of graphs known as banana graphs, with two marked vertices, through the lens of Hurwitz-Brill-Noether theory
  • As an application, we construct explicit new examples of finite graphs which are Brill-Noether general
  • These are the first such examples since the analysis of chains of loops by Cools, Draisma, Payne and Robeva.

We analyze a family of graphs known as banana graphs, with two marked vertices, through the lens of Hurwitz-Brill-Noether theory. As an application, we construct explicit new examples of finite graphs which are Brill-Noether general. These are the first such examples since the analysis of chains of loops by Cools, Draisma, Payne and Robeva. The graphs constructed are chains of loops and "theta graphs," which are banana graphs of genus 2. We also demonstrate that almost all banana graphs of genus at least 3 cannot be used for this purpose, due either to failure of a submodularity condition or to the presence of far too many inversions in certain permutations associated to divisors called transmission permutations.

Authors: Nathan Pflueger, Noah Solomon.

2022-11-30

CREPE: Open-Domain Question Answering with False Presuppositions

  • Most existing question answering (QA) datasets, in contrast, assume all questions have well defined answers
  • This is in large part due to difficulty in retrieving relevant evidence passages from a large text corpus.
Information seeking users often pose questions with false presuppositions, especially when asking about unfamiliar topics. Most existing question answering (QA) datasets, in contrast, assume all questions have well defined answers. We introduce CREPE, a QA dataset containing a natural distribution of presupposition failures from online information-seeking forums. We find that 25% of questions contain false presuppositions, and provide annotations for these presuppositions and their corrections. Through extensive baseline experiments, we show that adaptations of existing open-domain QA models can find presuppositions moderately well, but struggle when predicting whether a presupposition is factually correct. This is in large part due to difficulty in retrieving relevant evidence passages from a large text corpus. CREPE provides a benchmark to study question answering in the wild, and our analyses provide avenues for future work in better modeling and further studying the task.

Authors: Xinyan Velocity Yu, Sewon Min, Luke Zettlemoyer, Hannaneh Hajishirzi.

2022-11-30

CLIPascene: Scene Sketching with Different Types and Levels of Abstraction

  • We distinguish between two types of abstraction
  • The second is defined by the visual simplicity of the sketch, moving from a detailed depiction to a sparse sketch
  • The first network learns the desired placement of strokes, while the second network learns to gradually remove strokes from the sketch without harming its recognizability and semantics.

In this paper, we present a method for converting a given scene image into a sketch using different types and multiple levels of abstraction. We distinguish between two types of abstraction. The first considers the fidelity of the sketch, varying its representation from a more precise portrayal of the input to a looser depiction. The second is defined by the visual simplicity of the sketch, moving from a detailed depiction to a sparse sketch. Using an explicit disentanglement into two abstraction axes -- and multiple levels for each one -- provides users additional control over selecting the desired sketch based on their personal goals and preferences. To form a sketch at a given level of fidelity and simplification, we train two MLP networks. The first network learns the desired placement of strokes, while the second network learns to gradually remove strokes from the sketch without harming its recognizability and semantics. Our approach is able to generate sketches of complex scenes including those with complex backgrounds (e.g., natural and urban settings) and subjects (e.g., animals and people) while depicting gradual abstractions of the input scene in terms of fidelity and simplicity.

Authors: Yael Vinker, Yuval Alaluf, Daniel Cohen-Or, Ariel Shamir.

2022-11-30

The Reflected Entanglement Spectrum for Free Fermions

  • We find the spectrum numerically and analytically in certain limits
  • For intervals that almost touch the reflected entanglement spectrum approaches the spectrum of the thermal density matrix.
We consider the reflected entropy and the associated entanglement spectrum for free fermions reduced to two intervals in 1+1 dimensions. Working directly in the continuum theory the reflected entropy can be extracted from the spectrum of a singular integral equation whose kernel is determined by the known free fermion modular evolved correlation function. We find the spectrum numerically and analytically in certain limits. For intervals that almost touch the reflected entanglement spectrum approaches the spectrum of the thermal density matrix. This suggests that the reflected entanglement spectrum is well suited to the task of extracting physical data of the theory directly from the ground state wave function.

Authors: Souvik Dutta, Thomas Faulkner, Simon Lin.

2022-11-30

The Hierarchy Problem and the Top Yukawa: An Alternative to Top Partner Solutions

  • The lack of evidence for such colored partners however drives these models into more and more fine-tuned regions
  • Here, an alternative means to mitigate the top loop, allowing for natural electroweak symmetry breaking, is presented
  • We first discuss possible effects via a modification of the running of the top Yukawa coupling
  • Then, we turn to models where the top Yukawa is generated at one-loop level
  • Originated from a dimension-six operator, it drops when crossing the mass threshold of new degrees of freedom
  • In either case, the top partners are replaced by some new top-philic particles with strong interaction.

We discuss the role of the top-quark Yukawa coupling $y_t$ concerning the hierarchy problem and construct an alternative scheme to the conventional solutions with top partners. In traditional models, like SUSY or composite Higgs, top partners cancel the top loop contribution to the Higgs quadratic term. The lack of evidence for such colored partners however drives these models into more and more fine-tuned regions. Here, an alternative means to mitigate the top loop, allowing for natural electroweak symmetry breaking, is presented. Emphasizing that we have not measured the top-Higgs interactions at high scales yet, we envisage scenarios where this interaction is only approaching its sizable strength in the infra-red, but gets strongly suppressed at high scales. We first discuss possible effects via a modification of the running of the top Yukawa coupling. Then, we turn to models where the top Yukawa is generated at one-loop level. Originated from a dimension-six operator, it drops when crossing the mass threshold of new degrees of freedom. In either case, the top partners are replaced by some new top-philic particles with strong interaction. Thus, a very different phenomenology, such as large top mass running and signals in four top final states, is introduced, which will be discussed in detail. With the assistance of this mechanism, the solution to the hierarchy problem can be pushed to a (well-defined) higher scale, and a final test of naturalness might be deferred to a 100 TeV Collider, like the FCC.

Authors: Andreas Bally, Yi Chung, Florian Goertz.

2022-11-30

Definitions of entwinement

  • The literature now contains different, and sometimes inequivalent, field theory definitions of entwinement
  • We discuss similarities and differences with previous definitions of entwinement.
Entwinement was first introduced as the CFT dual to extremal, non-minimal geodesics of quotiented AdS$_3$ spaces. It was heuristically meant to capture the entanglement of internal, gauged degrees of freedom, for instance in the symmetric product orbifold CFT of the D1/D5 brane system. The literature now contains different, and sometimes inequivalent, field theory definitions of entwinement. In this paper, we build a discretized lattice model of symmetric product orbifold CFTs, and explicitly construct a gauge-invariant reduced density matrix whose von Neumann entropy agrees with the holographic computation of entwinement. Refining earlier notions, our construction gives meaning to the entwinement of an interval of given size within a long string of specific length. We discuss similarities and differences with previous definitions of entwinement.

Authors: Ben Craps, Marine De Clerck, Alejandro Vilar López.

2022-11-30

Mixed Valence Pseudobrookite Al1.75Ti1.25O5: High Temperature Phase Transitions, Magnetism and Resistivity

  • Like the known aluminum titanate pseudobrookites, anistropic thermal expansion is observed.

Dark blue single crystals of Al_1.75^(3+) Ti_1.0^(4+) Ti_0.25^(3+) O_5 were grown with a novel synthesis method based on the reaction of a Ti3+/Ti4+ containing langbeinite melt and Al2O3. The obtained needles crystallize in the pseudobrookite structure and undergo two reversible phase transitions from orthorhombic Cmcm to C2/m first and subsequently to C2 symmetry. Like the known aluminum titanate pseudobrookites, anistropic thermal expansion is observed. The temperature evolution of the crystal structure reveals some insights into the mechanism leading to the decomposition of the Al1.75Ti1.25O5 above 725{\deg}C. The magnetic and electrical properties are discussed and compared to other reported aluminum titanate pseudobrookites.

Authors: Davor Tolj, WenHua Bi, Yong Liu, Ivica Zivkovic, Henrik M. Ronnow, Arnaud Magrez.

2022-11-30

Quantum fluctuations in one-dimensional supersolids

  • In one-dimension, quantum fluctuations prevent the appearance of long-range order in a supersolid, and only quasi long-range order can survive
  • The analysis is based on an effective low-energy description accounting for the two coupled Goldstone modes.
In one-dimension, quantum fluctuations prevent the appearance of long-range order in a supersolid, and only quasi long-range order can survive. We derive this quantum critical behavior and study its influence on the superfluid response and properties of the solid. The analysis is based on an effective low-energy description accounting for the two coupled Goldstone modes. We find that the quantum phase transition from the superfluid to the supersolid is shifted by quantum fluctuations from its mean-field prediction. However, for current experimental parameters with dipolar atomic gases, this shift is not observable and the transition appears to be mean-field like.

Authors: Chris Bühler, Tobias Ilg, Hans Peter Büchler.

2022-11-30

Safe Model-Free Reinforcement Learning using Disturbance-Observer-Based Control Barrier Functions

  • Safe reinforcement learning (RL) with assured satisfaction of hard state constraints during training has recently received a lot of attention
  • Existing safety filter-based approaches typically involve learning of uncertain dynamics and quantifying the learned model error, which leads to conservative filters before a large amount of data is collected to learn a good model, thereby preventing efficient exploration
  • This paper presents a method for safe and efficient model-free RL using disturbance observers (DOBs) and control barrier functions (CBFs)
  • The DOB-based CBF can be used as a safety filter with any model-free RL algorithms by minimally modifying the actions of an RL agent whenever necessary to ensure safety throughout the learning process.

Safe reinforcement learning (RL) with assured satisfaction of hard state constraints during training has recently received a lot of attention. Safety filters, e.g., based on control barrier functions (CBFs), provide a promising way for safe RL via modifying the unsafe actions of an RL agent on the fly. Existing safety filter-based approaches typically involve learning of uncertain dynamics and quantifying the learned model error, which leads to conservative filters before a large amount of data is collected to learn a good model, thereby preventing efficient exploration. This paper presents a method for safe and efficient model-free RL using disturbance observers (DOBs) and control barrier functions (CBFs). Unlike most existing safe RL methods that deal with hard state constraints, our method does not involve model learning, and leverages DOBs to accurately estimate the pointwise value of the uncertainty, which is then incorporated into a robust CBF condition to generate safe actions. The DOB-based CBF can be used as a safety filter with any model-free RL algorithms by minimally modifying the actions of an RL agent whenever necessary to ensure safety throughout the learning process. Simulation results on a unicycle and a 2D quadrotor demonstrate that the proposed method outperforms a state-of-the-art safe RL algorithm using CBFs and Gaussian processes-based model learning, in terms of safety violation rate, and sample and computational efficiency.

Authors: Yikun Cheng, Pan Zhao, Naira Hovakimyan.