Search

We investigate the nonlinear stability of global Anti-de Sitter space in the presence of multitrace deformations utilizing an Einstein-Klein-Gordon system with a top-down scalar potential. Our numerical simulations show that marginal and irrelevant deformations retain the nonlinear instability originally found by Bizo\'n and Rostworowski, while relevant deformations disrupt it. We show that the nonlinear impact of multitrace deformations is in one-to-one correspondence with how they affect the resonant character of the global Anti-de Sitter normal mode eigenfrequency spectrum., Comment: 9 pages + appendices, 11 figures

When evaluating stimuli reconstruction results it is tempting to assume that higher fidelity text and image generation is due to an improved understanding of the brain or more powerful signal extraction from neural recordings. However, in practice, new reconstruction methods could improve performance for at least three other reasons: learning more about the distribution of stimuli, becoming better at reconstructing text or images in general, or exploiting weaknesses in current image and/or text evaluation metrics. Here we disentangle how much of the reconstruction is due to these other factors vs. productively using the neural recordings. We introduce BrainBits, a method that uses a bottleneck to quantify the amount of signal extracted from neural recordings that is actually necessary to reproduce a method's reconstruction fidelity. We find that it takes surprisingly little information from the brain to produce reconstructions with high fidelity. In these cases, it is clear that the priors of the methods' generative models are so powerful that the outputs they produce extrapolate far beyond the neural signal they decode. Given that reconstructing stimuli can be improved independently by either improving signal extraction from the brain or by building more powerful generative models, improving the latter may fool us into thinking we are improving the former. We propose that methods should report a method-specific random baseline, a reconstruction ceiling, and a curve of performance as a function of bottleneck size, with the ultimate goal of using more of the neural recordings., Comment: 23 pages, 16 figures, Accepted at NeurIPS 2024

We revisit the calculation of the soft gluon emission probability off a colour-singlet $q\bar q$ system that evolves in a quark-gluon plasma. The $q\bar q$ antenna is created in the presence of a medium and then emits a soft gluon outside. The gluon emission probability is modified with respect to the vacuum baseline due to interactions with the medium during the formation of the antenna and its propagation. Previous studies disregarded the former effect and found that the medium modification to the interference pattern of the antenna was controlled by the so-called critical angle $\theta_c$, that exclusively depends on medium properties. We find that accounting for medium interactions during the antenna formation enhances the total rate of emissions off the $q\bar q$ antenna. Interestingly, it also promotes the notion of a critical angle to a dynamic quantity, denoted $\tilde\theta_c$, that depends on both the medium and the antenna properties and is thus different for every splitting. As a consequence, depending on the region of parameter space, colour decoherence can either be delayed or accelerated with respect to previous estimates.

When we experience a visual stimulus as beautiful, how much of that experience derives from perceptual computations we cannot describe versus conceptual knowledge we can readily translate into natural language? Disentangling perception from language in visually-evoked affective and aesthetic experiences through behavioral paradigms or neuroimaging is often empirically intractable. Here, we circumnavigate this challenge by using linear decoding over the learned representations of unimodal vision, unimodal language, and multimodal (language-aligned) deep neural network (DNN) models to predict human beauty ratings of naturalistic images. We show that unimodal vision models (e.g. SimCLR) account for the vast majority of explainable variance in these ratings. Language-aligned vision models (e.g. SLIP) yield small gains relative to unimodal vision. Unimodal language models (e.g. GPT2) conditioned on visual embeddings to generate captions (via CLIPCap) yield no further gains. Caption embeddings alone yield less accurate predictions than image and caption embeddings combined (concatenated). Taken together, these results suggest that whatever words we may eventually find to describe our experience of beauty, the ineffable computations of feedforward perception may provide sufficient foundation for that experience.

Magnetic Resonance Fingerprinting (MRF) is a time-efficient approach to quantitative MRI, enabling the mapping of multiple tissue properties from a single, accelerated scan. However, achieving accurate reconstructions remains challenging, particularly in highly accelerated and undersampled acquisitions, which are crucial for reducing scan times. While deep learning techniques have advanced image reconstruction, the recent introduction of diffusion models offers new possibilities for imaging tasks, though their application in the medical field is still emerging. Notably, diffusion models have not yet been explored for the MRF problem. In this work, we propose for the first time a conditional diffusion probabilistic model for MRF image reconstruction. Qualitative and quantitative comparisons on in-vivo brain scan data demonstrate that the proposed approach can outperform established deep learning and compressed sensing algorithms for MRF reconstruction. Extensive ablation studies also explore strategies to improve computational efficiency of our approach., Comment: 12 pages, 5 figures, 2 algorithms

We demonstrate that architectures which traditionally are considered to be ill-suited for a task can be trained using inductive biases from another architecture. Networks are considered untrainable when they overfit, underfit, or converge to poor results even when tuning their hyperparameters. For example, plain fully connected networks overfit on object recognition while deep convolutional networks without residual connections underfit. The traditional answer is to change the architecture to impose some inductive bias, although what that bias is remains unknown. We introduce guidance, where a guide network guides a target network using a neural distance function. The target is optimized to perform well and to match its internal representations, layer-by-layer, to those of the guide; the guide is unchanged. If the guide is trained, this transfers over part of the architectural prior and knowledge of the guide to the target. If the guide is untrained, this transfers over only part of the architectural prior of the guide. In this manner, we can investigate what kinds of priors different architectures place on untrainable networks such as fully connected networks. We demonstrate that this method overcomes the immediate overfitting of fully connected networks on vision tasks, makes plain CNNs competitive to ResNets, closes much of the gap between plain vanilla RNNs and Transformers, and can even help Transformers learn tasks which RNNs can perform more easily. We also discover evidence that better initializations of fully connected networks likely exist to avoid overfitting. Our method provides a mathematical tool to investigate priors and architectures, and in the long term, may demystify the dark art of architecture creation, even perhaps turning architectures into a continuous optimizable parameter of the network., Comment: Under Review; 24 pages, 9 figures; Project page and code is at https://untrainable-networks.github.io/

This paper provides an overview of the techniques employed by Hibikino-Musashi@Home, which intends to participate in the domestic standard platform league. The team has developed a dataset generator for training a robot vision system and an open-source development environment running on a Human Support Robot simulator. The large language model powered task planner selects appropriate primitive skills to perform the task requested by users. The team aims to design a home service robot that can assist humans in their homes and continuously attends competitions to evaluate and improve the developed system.

Real-world applications may be affected by outlying values. In the model-based clustering literature, several methodologies have been proposed to detect units that deviate from the majority of the data (rowwise outliers) and trim them from the parameter estimates. However, the discarded observations can encompass valuable information in some observed features. Following the more recent cellwise contamination paradigm, we introduce a Gaussian mixture model for cellwise outlier detection. The proposal is estimated via an Expectation-Maximization (EM) algorithm with an additional step for flagging the contaminated cells of a data matrix and then imputing -- instead of discarding -- them before the parameter estimation. This procedure adheres to the spirit of the EM algorithm by treating the contaminated cells as missing values. We analyze the performance of the proposed model in comparison with other existing methodologies through a simulation study with different scenarios and illustrate its potential use for clustering, outlier detection, and imputation on three real data sets.

Magnetic Resonance Fingerprinting (MRF) is a time-efficient approach to quantitative MRI for multiparametric tissue mapping. The reconstruction of quantitative maps requires tailored algorithms for removing aliasing artefacts from the compressed sampled MRF acquisitions. Within approaches found in the literature, many focus solely on two-dimensional (2D) image reconstruction, neglecting the extension to volumetric (3D) scans despite their higher relevance and clinical value. A reason for this is that transitioning to 3D imaging without appropriate mitigations presents significant challenges, including increased computational cost and storage requirements, and the need for large amount of ground-truth (artefact-free) data for training. To address these issues, we introduce StoDIP, a new algorithm that extends the ground-truth-free Deep Image Prior (DIP) reconstruction to 3D MRF imaging. StoDIP employs memory-efficient stochastic updates across the multicoil MRF data, a carefully selected neural network architecture, as well as faster nonuniform FFT (NUFFT) transformations. This enables a faster convergence compared against a conventional DIP implementation without these features. Tested on a dataset of whole-brain scans from healthy volunteers, StoDIP demonstrated superior performance over the ground-truth-free reconstruction baselines, both quantitatively and qualitatively., Comment: 10 pages, 2 figures, 1 table, 1 algorithm

The estimation of multi-parametric quantitative maps from Magnetic Resonance Fingerprinting (MRF) compressed sampled acquisitions, albeit successful, remains a challenge due to the high underspampling rate and artifacts naturally occuring during image reconstruction. Whilst state-of-the-art DL methods can successfully address the task, to fully exploit their capabilities they often require training on a paired dataset, in an area where ground truth is seldom available. In this work, we propose a method that combines a deep image prior (DIP) module that, without ground truth and in conjunction with a Bloch consistency enforcing autoencoder, can tackle the problem, resulting in a method faster and of equivalent or better accuracy than DIP-MRF., Comment: 4 pages, 3 figures 1 table, presented at ISBI 2024

This paper considers generic scalar fields models in one space dimension,the potentials of which have several vacuum states. It is shown that the partial breaking of the vacuum degeneracy provides a mechanism for the appearance of static multi-kinks, whenever some false vacuum states appear between two true vacuum states. A multi-kink is a stable field configuration with an internal structure, consisting of various localized energy packages with well-defined separations between them. General results for multi-kinks properties are presented. We introduce the commensurable and non degenerate double sine-Gordon model as a specific realization of this mechanism, supporting the existence of multi-kinks with an arbitrary $n$-number of subkinks. The multi-kinks properties are thoroughly analyzed including the novel phenomenology displayed when two of them collide., Comment: 25 pages, 11 figures

We present optical spectroscopy of 710 solar neighborhood stars collected over twenty years to catalog chromospheric activity and search for stellar activity cycles. The California Legacy Survey stars are amenable to exoplanet detection using precise radial velocities, and we present their Ca II H and K time series as a proxy for stellar and chromospheric activity. Using the HIRES spectrometer at Keck Observatory, we measured stellar flux in the cores of the Ca II H and K lines to determine S-values on the Mt. Wilson scale and the log(R'HK) metric, which is comparable across a wide range of spectral types. From the 710 stars, with 52,372 observations, 285 stars are sufficiently sampled to search for stellar activity cycles with periods of 2-25 years, and 138 stars show stellar cycles of varying length and amplitude. S-values can be used to mitigate stellar activity in the detection and characterization of exoplanets. We use them to probe stellar dynamos and to place the Sun's magnetic activity into context among solar neighborhood stars. Using precise stellar parameters and time-averaged activity measurements, we find tightly constrained cycle periods as a function of stellar temperature between log(R'HK) of -4.7 and -4.9, a range of activity in which nearly every star has a periodic cycle. These observations present the largest sample of spectroscopically determined stellar activity cycles to date., Comment: 40 pages, 26 figures, submitted to ApJS

We establish the local Hadamard well-posedness of a certain third-order nonlinear Schr\"odinger equation with a multi-term linear part and a general power nonlinearity known as the higher-order nonlinear Schr\"odinger equation, formulated on a finite interval with a combination of nonzero Dirichlet and Neumann boundary conditions. Specifically, for initial and boundary data in suitable Sobolev spaces that are related to one another through the time regularity induced by the equation, we prove the existence of a unique solution as well as the continuous dependence of that solution on the data. The precise choice of solution space depends on the value of the Sobolev exponent and is dictated both by the linear estimates associated with the forced linear counterpart of the nonlinear initial-boundary value problem and, in the low-regularity setting below the Sobolev algebra property threshold, by certain nonlinear estimates that control the Sobolev norm of the power nonlinearity. In particular, as usual in Schr\"odinger-type equations, in the case of low regularity it is necessary to derive Strichartz estimates in suitable Lebesgue/Bessel potential spaces. The proof of well-posedness is based on a contraction mapping argument combined with the linear estimates, which are established by employing the explicit solution formula for the forced linear problem derived via the unified transform of Fokas. Due to the nature of the finite interval problem, this formula involves contour integrals in the complex Fourier plane with corresponding integrands that contain differences of exponentials in their denominators, thus requiring delicate handling through appropriate contour deformations. It is worth noting that, in addition to the various linear and nonlinear results obtained for the finite interval problem, novel time regularity results are established here also for the relevant half-line problem.

We compute the spectrum of gluons emitted by a highly energetic quark inside a flowing QCD medium, focusing on the soft gluon limit at first order in the opacity expansion. Specifically, we derive the leading energy-suppressed corrections to the double differential final parton distribution, and show that they are substantial for both static and flowing matter. In particular, we demonstrate that the corrections due to the transverse flow become large even at moderate energies and flow velocities, affecting drastically both the shape and magnitude of the spectrum. We observe that the final transverse momentum of the emitted gluon tends to align along the flow direction, resulting in a non-trivial azimuthal distribution. These results can be directly implemented into the estimation of multiple observables to get a better understanding of the jet-medium interaction processes in HICs., Comment: 29 pages, 5 figures

Inspired by the recent considerations of parton momentum broadening in the glasma phase, we study the medium-induced soft gluon radiation of jet partons at early times in heavy-ion collisions. The glasma state is assumed to be comprised of independent color domains with homogenous longitudinal fields that vary event by event, and we further complete this model with an event-averaging procedure accounting for the finite correlation length. Using this description, we evaluate the rate of medium-induced radiation from an energetic parton at midrapidity in the glasma phase. We mainly focus on SU(2) color fields for simplicity, also referring to the U(1) case and comparing with the BDMPS-Z rate to gain further insight. Our results show that there is an intricate interplay of the synchrotron-like radiation in a single color domain with the destructive interference between different color domains, after the medium averaging is performed. Thus, we find that the emission rate is sensitive to the matter structure, decreasing for a glasma state populated by smaller color domains, i.e. for a glasma with a larger characteristic saturation scale. Our approach can be applied to more realistic backgrounds, and sets the stage for the modelling of jet evolution in the early stages of heavy-ion collisions., Comment: 30 pages, 6 figures

We introduce a computationally efficient algorithm for zeroth-order bandit convex optimisation and prove that in the adversarial setting its regret is at most $d^{3.5} \sqrt{n} \mathrm{polylog}(n, d)$ with high probability where $d$ is the dimension and $n$ is the time horizon. In the stochastic setting the bound improves to $M d^{2} \sqrt{n} \mathrm{polylog}(n, d)$ where $M \in [d^{-1/2}, d^{-1 / 4}]$ is a constant that depends on the geometry of the constraint set and the desired computational properties.

We study the impact of entangling measurements on the efficiency of quantum measurement- based engines. We first show that for engines comprising many subsystems their efficiency can be enhanced by performing entangling measurements, as opposed to local measurements over each subsystem. When the collective measurement produces the same local state for the subsystems as individual local measurements, the improvement in the efficiency is proportional to the amount of correlations. Finally, we show that for two level systems these type of engine can operate at perfect efficiency while yielding a finite amount of work, in the limit large the number of subsystems., Comment: 6 pages, 2 figures

The Transiting Exoplanet Survey Satellite (TESS) has discovered hundreds of new worlds, with TESS planet candidates now outnumbering the total number of confirmed planets from $\textit{Kepler}$. Owing to differences in survey design, TESS continues to provide planets that are better suited for subsequent follow-up studies, including mass measurement through radial velocity (RV) observations, compared to Kepler targets. In this work, we present the TESS-Keck Survey's (TKS) Mass Catalog: a uniform analysis of all TKS RV survey data which has resulted in mass constraints for 126 planets and candidate signals. This includes 58 mass measurements that have reached $\geq5\sigma$ precision. We confirm or validate 32 new planets from the TESS mission either by significant mass measurement (15) or statistical validation (17), and we find no evidence of likely false positives among our entire sample. This work also serves as a data release for all previously unpublished TKS survey data, including 9,204 RV measurements and associated activity indicators over our three year survey. We took the opportunity to assess the performance of our survey, and found that we achieved many of our goals including measuring the mass of 38 small ($<4R_{\oplus}$) planets, nearly achieving the TESS mission's basic science requirement. In addition, we evaluated the performance of the Automated Planet Finder (APF) as survey support and observed meaningful constraints on system parameters due to its more uniform phase coverage. Finally, we compared our measured masses to those predicted by commonly used mass-radius relations and investigated evidence of systematic bias., Comment: 51 pages (22 of text), 24 figures

Chiral magnetic textures give rise to unconventional magnetotransport phenomena such as the topological Hall effect and nonreciprocal electronic transport. While the correspondence between real-space magnetic topology/symmetry and such transport phenomena has been well established, a microscopic understanding based on the spin-dependent band structure in momentum space remains elusive. Here we demonstrate how a chiral magnetic structure in real space introduces an asymmetry in the electronic band structure and triggers a nonreciprocal electronic transport in a centrosymmetric helimagnet {\alpha}-EuP$_3$. The magnetic structure of {\alpha}-EuP$_3$ is highly tunable by a magnetic field and closely coupled to its semi-metallic electronic band structure, enabling a systematic study across chiral and achiral magnetic phases on the correspondence between nonreciprocal transport and electronic band asymmetry. Our findings reveal how a microscopic change in the magnetic configuration of charge carriers can lead to nonreciprocal electronic transport, paving the way for designing chiral magnets with desirable properties., Comment: 31 pages, 12 figures, 1 table

The free energy of ion solvation can be decomposed into enthalpic and entropic contributions. This helps to understand the connection between the dielectric properties and the underlying forces. We present a simple linear-response model of screened charge interactions that provides an alternative understanding of solvation barriers. Moreover, it explains the "anti-Coulomb" interactions (attraction between like-charged ions and repulsion between opposite-charged ions) observed in both simulations and experiments. We show that this is a universal behavior associated to the non-local response function of any dielectric or metallic system., Comment: 10 + 9 pages, 8 + 14 figures

Social media data has been used for detecting users with mental disorders, such as depression. Despite the global significance of cross-cultural representation and its potential impact on model performance, publicly available datasets often lack crucial metadata related to this aspect. In this work, we evaluate the generalization of benchmark datasets to build AI models on cross-cultural Twitter data. We gather a custom geo-located Twitter dataset of depressed users from seven countries as a test dataset. Our results show that depression detection models do not generalize globally. The models perform worse on Global South users compared to Global North. Pre-trained language models achieve the best generalization compared to Logistic Regression, though still show significant gaps in performance on depressed and non-Western users. We quantify our findings and provide several actionable suggestions to mitigate this issue., Comment: 6 pages, 2 figures, NAACL 2024 Main Conference

We collected near-infrared spectra of 65 cool stars with the NASA InfraRed Telescope Facility (IRTF) and analyze them to calculate accurate metallicities and stellar parameters. The sample of 55 M dwarfs and 10 K dwarfs includes 25 systems with confirmed planets and 27 systems with planet candidates identified by the K2 and TESS missions. Three of the 25 confirmed planetary systems host multiple confirmed planets and two of the 27 planet candidate systems host multiple planet candidates. Using the new stellar parameters, we re-fit the K2 and TESS light curves to calculate updated planet properties. In general, our updated stellar properties are more precise than those previously reported and our updated planet properties agree well with those in the literature. Lastly, we briefly examine the relationship between stellar mass, stellar metallicity, and planetary system properties for targets in our sample and for previously characterized planet-hosting low-mass stars. We provide our spectra, stellar parameters, and new planetary fits to the community, expanding the sample available with which to investigate correlations between stellar and planetary properties for low-mass stars., Comment: 4 tables, 6 figures

In 2019, the paper entitled "Model Cards for Model Reporting" introduced a new tool for documenting model performance and encouraged the practice of transparent reporting for a defined list of categories. One of the categories detailed in that paper is ethical considerations, which includes the subcategories of data, human life, mitigations, risks and harms, and use cases. We propose to reclassify this category in the original model card due to the recent maturing of the field known as trustworthy AI, a term which analyzes whether the algorithmic properties of the model indicate that the AI system is deserving of trust from its stakeholders. In our examination of trustworthy AI, we highlight three respected organizations - the European Commission's High-Level Expert Group on AI, the OECD, and the U.S.-based NIST - that have written guidelines on various aspects of trustworthy AI. These recent publications converge on numerous characteristics of the term, including accountability, explainability, fairness, privacy, reliability, robustness, safety, security, and transparency, while recognizing that the implementation of trustworthy AI varies by context. Our reclassification of the original model-card category known as ethical considerations involves a two-step process: 1) adding a new category known as trustworthiness, where the subcategories will be derived from the discussion of trustworthy AI in our paper, and 2) maintaining the subcategories of ethical considerations under a renamed category known as risk environment and risk management, a title which we believe better captures today's understanding of the essence of these topics. We hope that this reclassification will further the goals of the original paper and continue to prompt those releasing trained models to accompany these models with documentation that will assist in the evaluation of their algorithmic properties., Comment: 14 pages, 2 figures, submitted to ACM Conference on Fairness, Accountability, and Transparency 2024 (ACM FAccT '24)

The extreme environments of ultra-short-period planets (USPs) make excellent laboratories to study how exoplanets obtain, lose, retain, and/or regain gaseous atmospheres. We present the confirmation and characterization of the USP TOI-1347 b, a $1.8 \pm 0.1$ R$_\oplus$ planet on a 0.85 day orbit that was detected with photometry from the TESS mission. We measured radial velocities of the TOI-1347 system using Keck/HIRES and HARPS-N and found the USP to be unusually massive at $11.1 \pm 1.2$ M$_\oplus$. The measured mass and radius of TOI-1347 b imply an Earth-like bulk composition. A thin H/He envelope (>0.01% by mass) can be ruled out at high confidence. The system is between 1 and 1.8 Gyr old; therefore, intensive photoevaporation should have concluded. We detected a tentative phase curve variation (3$\sigma$) and a secondary eclipse (2$\sigma$) in TESS photometry, which if confirmed could indicate the presence of a high-mean-molecular-weight atmosphere. We recommend additional optical and infrared observations to confirm the presence of an atmosphere and investigate its composition., Comment: 22 pages, 9 figures, accepted for publication in The Astronomical Journal

We present and confirm TOI-1751 b, a transiting sub-Neptune orbiting a slightly evolved, solar-type, metal-poor star ($T_{eff} = 5996 \pm 110$ K, $log(g) = 4.2 \pm 0.1$, V = 9.3 mag, [Fe/H] = $-0.40 \pm 0.06$ dex) every 37.47 d. We use TESS photometry to measure a planet radius of $2.77_{-0.07}^{+0.15}~\rm{R_\oplus}$. We also use both Keck/HIRES and APF/Levy radial velocities (RV) to derive a planet mass of $14.5_{-3.14}^{+3.15} ~\rm{M_\oplus}$, and thus a planet density of $3.6 \pm 0.9 \, {\rm g}\,{\rm cm}^{-3}$. There is also a long-period ($\sim400~\rm{d}$) signal that is observed in only the Keck/HIRES data. We conclude that this long-period signal is not planetary in nature, and is likely due to the window function of the Keck/HIRES observations. This highlights the role of complementary observations from multiple observatories to identify and exclude aliases in RV data. Finally, we investigate potential compositions of this planet, including rocky and water-rich solutions, as well as theoretical irradiated ocean models. TOI-1751 b is a warm sub-Neptune, with an equilibrium temperature of $\sim 820$ K. As TOI-1751 is a metal-poor star, TOI-1751 b may have formed in a water-enriched formation environment. We thus favor a volatile-rich interior composition for this planet., Comment: 30 pages, 9 figures. Accepted for publication in AJ