Search

Masked language modelling (MLM) as a pretraining objective has been widely adopted in genomic sequence modelling. While pretrained models can successfully serve as encoders for various downstream tasks, the distribution shift between pretraining and inference detrimentally impacts performance, as the pretraining task is to map [MASK] tokens to predictions, yet the [MASK] is absent during downstream applications. This means the encoder does not prioritize its encodings of non-[MASK] tokens, and expends parameters and compute on work only relevant to the MLM task, despite this being irrelevant at deployment time. In this work, we propose a modified encoder-decoder architecture based on the masked autoencoder framework, designed to address this inefficiency within a BERT-based transformer. We empirically show that the resulting mismatch is particularly detrimental in genomic pipelines where models are often used for feature extraction without fine-tuning. We evaluate our approach on the BIOSCAN-5M dataset, comprising over 2 million unique DNA barcodes. We achieve substantial performance gains in both closed-world and open-world classification tasks when compared against causal models and bidirectional architectures pretrained with MLM tasks., Comment: 10 pages, 5 figures

Vision foundation models (VFMs) are pre-trained on extensive image datasets to learn general representations for diverse types of data. These models can subsequently be fine-tuned for specific downstream tasks, significantly boosting performance across a broad range of applications. However, existing vision foundation models that claim to be applicable to various clinical tasks are mostly pre-trained on 3D computed tomography (CT), which benefits from the availability of extensive 3D CT databases. Significant differences between CT and magnetic resonance imaging (MRI) in imaging principles, signal characteristics, and data distribution may hinder their practical performance and versatility in MRI-specific applications. Here, we propose Triad, a vision foundation model for 3D MRI. Triad adopts a widely used autoencoder architecture to learn robust representations from 131,170 3D MRI volumes and uses organ-independent imaging descriptions to constrain the semantic distribution of the visual modality. The above pre-training dataset is called Triad-131K, which is currently the largest 3D MRI pre-training dataset. We evaluate Triad across three tasks, namely, organ/tumor segmentation, organ/cancer classification, and medical image registration, in two data modalities (within-domain and out-of-domain) settings using 25 downstream datasets. By initializing models with Triad's pre-trained weights, nnUNet-Triad improves segmentation performance by 2.51% compared to nnUNet-Scratch across 17 datasets. Swin-B-Triad achieves a 3.97% improvement over Swin-B-Scratch in classification tasks across five datasets. SwinUNETR-Triad improves by 4.00% compared to SwinUNETR-Scratch in registration tasks across two datasets. Our study demonstrates that pre-training can improve performance when the data modalities and organs of upstream and downstream tasks are consistent.

Background: MRI is crucial for brain imaging but is highly susceptible to motion artifacts due to long acquisition times. This study introduces PI-MoCoNet, a physics-informed motion correction network that integrates spatial and k-space information to remove motion artifacts without explicit motion parameter estimation, enhancing image fidelity and diagnostic reliability. Materials and Methods: PI-MoCoNet consists of a motion detection network (U-net with spatial averaging) to identify corrupted k-space lines and a motion correction network (U-net with Swin Transformer blocks) to reconstruct motion-free images. The correction is guided by three loss functions: reconstruction (L1), perceptual (LPIPS), and data consistency (Ldc). Motion artifacts were simulated via rigid phase encoding perturbations and evaluated on IXI and MR-ART datasets against Pix2Pix, CycleGAN, and U-net using PSNR, SSIM, and NMSE. Results: PI-MoCoNet significantly improved image quality. On IXI, for minor artifacts, PSNR increased from 34.15 dB to 45.95 dB, SSIM from 0.87 to 1.00, and NMSE reduced from 0.55% to 0.04%. For moderate artifacts, PSNR improved from 30.23 dB to 42.16 dB, SSIM from 0.80 to 0.99, and NMSE from 1.32% to 0.09%. For heavy artifacts, PSNR rose from 27.99 dB to 36.01 dB, SSIM from 0.75 to 0.97, and NMSE decreased from 2.21% to 0.36%. On MR-ART, PI-MoCoNet achieved PSNR gains of ~10 dB and SSIM improvements of up to 0.20, with NMSE reductions of ~6%. Ablation studies confirmed the importance of data consistency and perceptual losses, yielding a 1 dB PSNR gain and 0.17% NMSE reduction. Conclusions: PI-MoCoNet effectively mitigates motion artifacts in brain MRI, outperforming existing methods. Its ability to integrate spatial and k-space information makes it a promising tool for clinical use in motion-prone settings. Code: https://github.com/mosaf/PI-MoCoNet.git.

Activation functions are fundamental elements of deep learning architectures as they significantly influence training dynamics. ReLU, while widely used, is prone to the dying neuron problem, which has been mitigated by variants such as LeakyReLU, PReLU, and ELU that better handle negative neuron outputs. Recently, self-gated activations like GELU and Swish have emerged as state-of-the-art alternatives, leveraging their smoothness to ensure stable gradient flow and prevent neuron inactivity. In this work, we introduce the Gompertz Linear Unit (GoLU), a novel self-gated activation function defined as $\mathrm{GoLU}(x) = x \, \mathrm{Gompertz}(x)$, where $\mathrm{Gompertz}(x) = e^{-e^{-x}}$. The GoLU activation leverages the asymmetry in the Gompertz function to reduce variance in the latent space more effectively compared to GELU and Swish, while preserving robust gradient flow. Extensive experiments across diverse tasks, including Image Classification, Language Modeling, Semantic Segmentation, Object Detection, Instance Segmentation, and Diffusion, highlight GoLU's superior performance relative to state-of-the-art activation functions, establishing GoLU as a robust alternative to existing activation functions., Comment: 8 pages, excluding references and appendix

FPGAs have transformed digital design by enabling versatile and customizable solutions that balance performance and power efficiency, yielding them essential for today's diverse computing challenges. Research in the Netherlands, both in academia and industry, plays a major role in developing new innovative FPGA solutions. This survey presents the current landscape of FPGA innovation research in the Netherlands by delving into ongoing projects, advancements, and breakthroughs in the field. Focusing on recent research outcome (within the past 5 years), we have identified five key research areas: a) FPGA architecture, b) FPGA robustness, c) data center infrastructure and high-performance computing, d) programming models and tools, and e) applications. This survey provides in-depth insights beyond a mere snapshot of the current innovation research landscape by highlighting future research directions within each key area; these insights can serve as a foundational resource to inform potential national-level investments in FPGA technology.

Modeling plays a critical role in additive manufacturing (AM), enabling a deeper understanding of underlying processes. Parametric solutions for such models are of great importance, enabling the optimization of production processes and considerable cost reductions. However, the complexity of the problem and diversity of spatio-temporal scales involved in the process pose significant challenges for traditional numerical methods. Surrogate models offer a powerful alternative by accelerating simulations and facilitating real-time monitoring and control. The present study presents an operator learning approach that relies on the deep operator network (DeepONet) and physics-informed neural networks (PINN) to predict the three-dimensional temperature distribution during melting and consolidation in laser powder bed fusion (LPBF). Parametric solutions for both single-track and multi-track scenarios with respect to tool path are obtained. To address the challenges in obtaining parametric solutions for multi-track scenarios using DeepONet architecture, a sequential PINN approach is proposed to efficiently manage the increased training complexity inherent in those scenarios. The accuracy and consistency of the model are verified against finite-difference computations. The developed surrogate allows us to efficiently analyze the effect of scanning paths and laser parameters on the thermal history.

We empirically analyze the reversion of financial market trends with time horizons ranging from minutes to decades. The analysis covers equities, interest rates, currencies and commodities and combines 14 years of futures tick data, 30 years of daily futures prices, 330 years of monthly asset prices, and yearly financial data since medieval times. Across asset classes, we find that markets are in a trending regime on time scales that range from a few hours to a few years, while they are in a reversion regime on shorter and longer time scales. In the trending regime, weak trends tend to persist, which can be explained by herding behavior of investors. However, in this regime trends tend to revert before they become strong enough to be statistically significant, which can be interpreted as a return of asset prices to their intrinsic value. In the reversion regime, we find the opposite pattern: weak trends tend to revert, while those trends that become statistically significant tend to persist. Our results provide a set of empirical tests of theoretical models of financial markets. We interpret them in the light of a recently proposed lattice gas model, where the lattice represents the social network of traders, the gas molecules represent the shares of financial assets, and efficient markets correspond to the critical point. If this model is accurate, the lattice gas must be near this critical point on time scales from 1 hour to a few days, with a correlation time of a few years., Comment: 33 pages, 10 figures

This paper presents a novel streaming end-to-end target-speaker speech recognition that addresses two critical limitations in systems: the handling of noisy enrollment utterances and specific enrollment phrase requirements. This paper proposes a robust Target-Speaker Recurrent Neural Network Transducer (TS-RNNT) with dual attention mechanisms for contextual biasing and overlapping enrollment processing. The model incorporates a text decoder and attention mechanism specifically designed to extract relevant speaker characteristics from noisy, overlapping enrollment audio. Experimental results on a synthesized dataset demonstrate the model's resilience, maintaining a Word Error Rate (WER) of 16.44% even with overlapping enrollment at 5dB Signal-to-Interference Ratio (SIR), compared to conventional approaches that degrade to WERs above 75% under similar conditions. This significant performance improvement, coupled with the model's semi-text-dependent enrollment capabilities, represents a substantial advancement toward more practical and versatile voice-controlled devices.

Y$_2$Co$_3$ is a newly discovered antiferromagnetic (AFM) compound with distorted kagome layers. Previous investigations via bulk magnetization measurements suggested a complex noncollinear magnetic behavior, with magnetic moments primarily anti-aligned along the $b$ axis and some canting towards the $ac$ plane. In this study, we report the magnetic structure of Y$_2$Co$_3$ to be an A-type AFM structure with ferromagnetic (FM) interactions within the distorted kagome plane and an interplane antiferromagnetic interaction, as determined by single-crystal neutron diffraction. The magnetic moments align along the $b$ axis, with minimal canting towards the $c$ axis, at odds with the previous interpretation of bulk magnetization measurements. The magnetic moments on the two distinct Co sites are [0, -0.68(3), 0] $\mu_B$ and [0, 1.25(4), 0.07(1)] $\mu_B$. We attribute the previously reported "noncollinear" behavior to the considerable temperature dependence of itinerant AFM exchange interactions, induced by thermal contraction along the $b$ axis. Additionally, our examination of lattice constants through pressure studies reveals compensating effects on FM and AFM interactions, resulting in negligible pressure dependence of $T_\textrm{N}$.

Magnetic resonance imaging (MRI) is a non-invasive imaging modality and provides comprehensive anatomical and functional insights into the human body. However, its long acquisition times can lead to patient discomfort, motion artifacts, and limiting real-time applications. To address these challenges, strategies such as parallel imaging have been applied, which utilize multiple receiver coils to speed up the data acquisition process. Additionally, compressed sensing (CS) is a method that facilitates image reconstruction from sparse data, significantly reducing image acquisition time by minimizing the amount of data collection needed. Recently, deep learning (DL) has emerged as a powerful tool for improving MRI reconstruction. It has been integrated with parallel imaging and CS principles to achieve faster and more accurate MRI reconstructions. This review comprehensively examines DL-based techniques for MRI reconstruction. We categorize and discuss various DL-based methods, including end-to-end approaches, unrolled optimization, and federated learning, highlighting their potential benefits. Our systematic review highlights significant contributions and underscores the potential of DL in MRI reconstruction. Additionally, we summarize key results and trends in DL-based MRI reconstruction, including quantitative metrics, the dataset, acceleration factors, and the progress of and research interest in DL techniques over time. Finally, we discuss potential future directions and the importance of DL-based MRI reconstruction in advancing medical imaging. To facilitate further research in this area, we provide a GitHub repository that includes up-to-date DL-based MRI reconstruction publications and public datasets-https://github.com/mosaf/Awesome-DL-based-CS-MRI., Comment: arXiv admin note: substantial text overlap with arXiv:2405.00241

Ant Colony Optimization (ACO) is a well-known method inspired by the foraging behavior of ants and is extensively used to solve combinatorial optimization problems. In this paper, we first consider a general framework based on the concept of a construction graph - a graph associated with an instance of the optimization problem under study, where feasible solutions are represented by walks. We analyze the running time of this ACO variant, known as the Graph-based Ant System with time-dependent evaporation rate (GBAS/tdev), and prove that the algorithm's solution converges to the optimal solution of the problem with probability 1 for a slightly stronger evaporation rate function than was previously known. We then consider two time-dependent adaptations of Attiratanasunthron and Fakcharoenphol's $n$-ANT algorithm: $n$-ANT with time-dependent evaporation rate ($n$-ANT/tdev) and $n$-ANT with time-dependent lower pheromone bound ($n$-ANT/tdlb). We analyze both variants on the single destination shortest path problem (SDSP). Our results show that $n$-ANT/tdev has a super-polynomial time lower bound on the SDSP. In contrast, we show that $n$-ANT/tdlb achieves a polynomial time upper bound on this problem.

Classifying galaxies is an essential step for studying their structures and dynamics. Using GalaxyZoo2 (GZ2) fractions thresholds, we collect 545 and 11,735 samples in non-galaxy and galaxy classes, respectively. We compute the Zernike moments (ZMs) for GZ2 images, extracting unique and independent characteristics of galaxies. The uniqueness due to the orthogonality and completeness of Zernike polynomials, reconstruction of the original images with minimum errors, invariances (rotation, translation, and scaling), different block structures, and discriminant decision boundaries of ZMs' probability density functions for different order numbers indicate the capability of ZMs in describing galaxy features. We classify the GZ2 samples, firstly into the galaxies and non-galaxies and secondly, galaxies into spiral, elliptical, and odd objects (e.g., ring, lens, disturbed, irregular, merger, and dust lane). The two models include the support vector machine (SVM) and one-dimensional convolutional neural network (1D-CNN), which use ZMs, compared with the other three classification models of 2D-CNN, ResNet50, and VGG16 that apply the features from original images. We find the true skill statistic (TSS) greater than 0.86 for the SVM and 1D-CNN with ZMs for the oversampled galaxy-non-galaxy classifier. The SVM with ZMs model has a high-performance classification for galaxy and non-galaxy datasets. We show that the SVM with ZMs, 1D-CNN with ZMs, and VGG16 with vision transformer are high-performance (accuracy larger than 0.90 and TSS greater than 0.86) models for classifying the galaxies into spiral, elliptical, and odd objects. We conclude that these machine-learning algorithms are helpful tools for classifying galaxy images., Comment: 23 pages, 15, figures, The Astrophysical Journal Supplement Series

A flow control framework based on linear stability analysis is proposed focusing on reducing the aerodynamic drag due to dynamic stall through a finite-window temporal actuation. The methodology is applied on a periodically plunging SD7003 airfoil.Finite-time Lyapunov exponent (FTLE) fields reveal a saddle point near the airfoil leading edge, where a shear layer forms and feeds a dynamic stall vortex (DSV). A local stability analysis conducted at this saddle point identifies a Kelvin-Helmholtz instability, and the most unstable eigenvalue frequencies remain constant when the variation in the effective angle of attack is minimal. The findings from the FTLE fields and the stability analysis are used to inform the position and finite duty cycle of a periodic blowing and suction actuation applied in a wall-resolved large eddy simulation (LES). The present framework reduces the actuation duty cycle by 77.5% during the airfoil plunging motion, while maintaining the same performance as a continuous actuation throughout the entire cycle. The LES results demonstrate that disturbances from the stability-analysis-informed actuation modify the leading-edge dynamics, preventing the formation of the coherent DSV and significantly reducing the drag., Comment: 23 pages, 10 figures

Quantum imaging is an ever expanding research field, in which the aim is to exploit the quantum nature of light to enhance image reconstruction capabilities. Despite a number of successful demonstrations for quantum imaging, quantum microscopy still seems out of the range for practical applications, due to different physical and technical reasons. Here we propose an imaging method exploiting the quantum correlations of photon pairs and a scanning microscope to achieve fast, single mode quantum imaging. We first test our technique on a metal grating to estimate the resolution capabilities of our system. Moreover, we asses its potential in terms of the number of available independent pixels at full resolution compared to different quantum imaging approaches. Finally, we demonstrate scanning quantum microscopy of onion epithelial cells, paving the way towards scalable quantum microscopy for bio-physical applications. Our results, combined with the rapidly evolving photon-pair generation and detection technology towards the mid-infrared, could lead to an extension of quantum microscopy applications towards the mid-infrared, to access the molecular fingerprint region.

As well known, the cross sections of the resonance and Quasi-Elastic (QE) scattering off nucleons depend on quantities known as form factors that describe the nucleon structure. There are alternative approaches to determine the values of these non-perturbative quantities, some of them relying on the Neutral Current (NC) scattering of neutrinos off nucleons. In the presence of NC Non-Standard Interactions (NSI), such derivations must be revisited. The aim of the present paper is to discuss how information on NSI can be extracted by combining alternative approaches for deriving the form factors. We discuss how the KamLAND atmospheric neutrino data with $E_\nu<{\rm GeV}$ (used to determine the axial strange form factor $g_A^s$) can already constrain the axial NSI of $\nu_\tau$ with nucleons. We also argue that if the precision measurement of $\nu_\mu$ NC QE scattering establishes unexpectedly large vector strange form factor ({\it e.g.,} $F_1^s(Q^2)\sim 0.01$), it will be an indication for nonzero NSI coupling with $u$ and $d$ quarks ($\epsilon_{\mu\mu}^{Au/d}\sim 0.01$). We study the QE and resonance scattering cross sections of $\nu_\tau$ and $\nu_e$ off Argon and show that if their axial NSI is of the order of (but of course below) the present bounds, the deviation of QE cross sections from the SM prediction will be sizable and distinct from the uncertainties induced by the form factors., Comment: 33 pages, 7 figures

Local search is a widely used technique for tackling challenging optimization problems, offering simplicity and strong empirical performance across various problem domains. In this paper, we address the problem of scheduling a set of jobs on identical parallel machines with the objective of makespan minimization, by considering a local search neighborhood, called $k$-swap. A $k$-swap neighbor is obtained by interchanging the machine allocations of at most $k$ jobs scheduled on two machines. While local search algorithms often perform well in practice, they can exhibit poor worst-case performance. In our previous study, we showed that for $k \geq 3$, there exists an instance where the number of iterations required to converge to a local optimum is exponential in the number of jobs. Motivated by this discrepancy between theoretical worst-case bound and practical performance, we apply smoothed analysis to the $k$-swap local search. Smoothed analysis has emerged as a powerful framework for analyzing the behavior of algorithms, aiming to bridge the gap between poor worst-case and good empirical performance. In this paper, we show that the smoothed number of iterations required to find a local optimum with respect to the $k$-swap neighborhood is bounded by $O(m^2 \cdot n^{2k+2} \cdot \log m \cdot \phi)$, where $n$ and $m$ are the numbers of jobs and machines, respectively, and $\phi \geq 1$ is the perturbation parameter. The bound on the smoothed number of iterations demonstrates that the proposed lower bound reflects a pessimistic scenario which is rare in practice.

A generalised extraction procedure for magnetic interactions using effective Hamiltonians applicable to systems with more than two sites featuring local spins $S_i \geq 1$ is presented. To this end, closed, non-recursive expressions pertaining to chains of arbitrary equal spins are derived with the graphical method of angular momentum. The method is illustrated by extracting magnetic couplings from ab initio calculations on a [CaMn$_3^{(\text{IV})}$O$_4$] cubane. An extension to non-sequential coupling schemes is shown to be conducive to expressing additional symmetries of certain spin Hamiltonians.

Due to the saturation of the body ($m_b$) and surface ($M_S$) earthquake magnitudes, the moment magnitude ($M_W$) is a more convenient parameter for representing earthquake energies. We use the HRVD data, including 18,569 earthquakes recorded from 1976 January 1 to 2010 December 31, for five regions on the Earth (\textbf{Asia}-Oceania, Europe, Africa, North America, and South America). Applying the error propagation technique on the moment tensor, we estimate the standard error for $M_W$ showing a distribution that deviates from Gaussian errors which recommends using the Monte Carlo Markov chain (MCMC) to obtain model parameters instead of the least square approach. We investigate the linear relationship between the body and surface waves in small ($\le$ 6.1) and large ($>$6.1) magnitudes via MCMC. The slope varies from 0.811 to 1.565 ($m_b - M_W$) and 0.566 to 0.971 ($M_S - M_W$) across five regions. For 18,569 global earthquakes, we obtained the slope for $m_b \le 6.1$ ($m_b>$6.1) 0.863 (1.374), while we find the slope for $M_S\le$6.1 ($M_S >$ 6.1) 0.581 (0.921). These transition relationships of the magnitude moments are useful for increasing earthquake monitoring capacity.

Uncertainty around predictions from a model due to the finite size of the development sample has traditionally been approached using classical inferential techniques. The finite size of the sample will result in the discrepancy between the final model and the correct model that maps covariates to predicted risks. From a decision-theoretic perspective, this discrepancy might affect the subsequent treatment decisions, and thus is associated with utility loss. From this perspective, procuring more development data is associated in expected gain in the utility of using the model. In this work, we define the Expected Value of Sample Information (EVSI) as the expected gain in clinical utility, defined in net benefit (NB) terms in net true positive units, by procuring a further development sample of a given size. We propose a bootstrap-based algorithm for EVSI computations, and show its feasibility and face validity in a case study. Decision-theoretic metrics can complement classical inferential methods when designing studies that are aimed at developing risk prediction models.

This paper presents an approach for navigation and control in unmapped environments under input and state constraints using a composite control barrier function (CBF). We consider the scenario where real-time perception feedback (e.g., LiDAR) is used online to construct a local CBF that models local state constraints (e.g., local safety constraints such as obstacles) in the a priori unmapped environment. The approach employs a soft-maximum function to synthesize a single time-varying CBF from the N most recently obtained local CBFs. Next, the input constraints are transformed into controller-state constraints through the use of control dynamics. Then, we use a soft-minimum function to compose the input constraints with the time-varying CBF that models the a priori unmapped environment. This composition yields a single relaxed CBF, which is used in a constrained optimization to obtain an optimal control that satisfies the state and input constraints. The approach is validated through simulations of a nonholonomic ground robot that is equipped with LiDAR and navigates an unmapped environment. The robot successfully navigates the environment while avoiding the a priori unmapped obstacles and satisfying both speed and input constraints., Comment: Preprint submitted to 2025 American Control Conference (ACC). arXiv admin note: substantial text overlap with arXiv:2409.01458

Objective: Gadolinium-based contrast agents (GBCAs) are commonly used in MRI scans of patients with gliomas to enhance brain tumor characterization using T1-weighted (T1W) MRI. However, there is growing concern about GBCA toxicity. This study develops a deep-learning framework to generate T1-postcontrast (T1C) from pre-contrast multiparametric MRI. Approach: We propose the tumor-aware vision transformer (TA-ViT) model that predicts high-quality T1C images. The predicted tumor region is significantly improved (P < .001) by conditioning the transformer layers from predicted segmentation maps through adaptive layer norm zero mechanism. The predicted segmentation maps were generated with the multi-parametric residual (MPR) ViT model and transformed into a latent space to produce compressed, feature-rich representations. The TA-ViT model predicted T1C MRI images of 501 glioma cases. Selected patients were split into training (N=400), validation (N=50), and test (N=51) sets. Main Results: Both qualitative and quantitative results demonstrate that the TA-ViT model performs superior against the benchmark MRP-ViT model. Our method produces synthetic T1C MRI with high soft tissue contrast and more accurately reconstructs both the tumor and whole brain volumes. The synthesized T1C images achieved remarkable improvements in both tumor and healthy tissue regions compared to the MRP-ViT model. For healthy tissue and tumor regions, the results were as follows: NMSE: 8.53 +/- 4.61E-4; PSNR: 31.2 +/- 2.2; NCC: 0.908 +/- .041 and NMSE: 1.22 +/- 1.27E-4, PSNR: 41.3 +/- 4.7, and NCC: 0.879 +/- 0.042, respectively. Significance: The proposed method generates synthetic T1C images that closely resemble real T1C images. Future development and application of this approach may enable contrast-agent-free MRI for brain tumor patients, eliminating the risk of GBCA toxicity and simplifying the MRI scan protocol., Comment: arXiv admin note: text overlap with arXiv:2407.02616

We present a closed-form optimal feedback control method that ensures safety in an a prior unknown and potentially dynamic environment. This article considers the scenario where local perception data (e.g., LiDAR) is obtained periodically, and this data can be used to construct a local control barrier function (CBF) that models a local set that is safe for a period of time into the future. Then, we use a smooth time-varying soft-maximum function to compose the N most recently obtained local CBFs into a single barrier function that models an approximate union of the N most recently obtained local sets. This composite barrier function is used in a constrained quadratic optimization, which is solved in closed form to obtain a safe-and-optimal feedback control. We also apply the time-varying soft-maximum barrier function control to 2 robotic systems (nonholonomic ground robot with nonnegligible inertia, and quadrotor robot), where the objective is to navigate an a priori unknown environment safely and reach a target destination. In these applications, we present a simple approach to generate local CBFs from periodically obtained perception data., Comment: Preprint submitted to IEEE Transactions on Control Systems Technology (TCST)

Different activation functions work best for different deep learning models. To exploit this, we leverage recent advancements in gradient-based search techniques for neural architectures to efficiently identify high-performing activation functions for a given application. We propose a fine-grained search cell that combines basic mathematical operations to model activation functions, allowing for the exploration of novel activations. Our approach enables the identification of specialized activations, leading to improved performance in every model we tried, from image classification to language models. Moreover, the identified activations exhibit strong transferability to larger models of the same type, as well as new datasets. Importantly, our automated process for creating customized activation functions is orders of magnitude more efficient than previous approaches. It can easily be applied on top of arbitrary deep learning pipelines and thus offers a promising practical avenue for enhancing deep learning architectures., Comment: 10 pages, 1 figure, excluding references and appendix

The differential rotation plays a crucial role in the dynamics of the Sun. We study the solar rotation and its correlation with solar activity by applying a modified machine learning algorithm to identify and track coronal bright points (CBPs) from the Solar Dynamics Observatory/Atmospheric Imaging Assembly observations at 193 \AA\ during cycle 24. For more than 321,440 CBPs, the sidereal and meridional velocities are computed. We find the occurring height of CBPs about 5627 km above the photosphere. We obtain a rotational map for the corona by tracking CBPs at the formation height of Fe\,{\sc xii} (193 \AA) emissions. The equator rotation (14.$^{\circ}$40 to 14.$^{\circ}$54 day$^{-1}$) and latitudinal gradient of rotation ($ - $3.$^{\circ}$0 to $ - $2.$^{\circ}$64 day$^{-1}$) show very slightly positive and negative trends with solar activity (sunspots and flares), respectively. For cycle 24, our investigations show that the northern hemisphere has more differential rotation than the southern hemisphere, confirmed by the asymmetry of the midlatitude rotation parameter. The asymmetry (ranked) of the latitudinal gradient of the rotation parameter is concordant with the sunspot numbers for 7 yr within the 9 yr of the cycle; however, for only 3 yr, it is concordant with the flare index. The minimum horizontal Reynolds stress changes from about $ - $2500 m$^{2}$ s$^{-2}$ (corresponding to high activity) in 2012 and 2014 to $ - $100 m$^{2}$ s$^{-2}$ (corresponding to low activity) in 2019 over 5$^{\circ}$ to 35$^{\circ}$ latitudes within cycle 24. We conclude that the negative horizontal Reynolds stress (momentum transfer toward the Sun's equator) is a helpful indication of solar activity.

The regular behavior of the pulsations of high-amplitude Delta Sct (HADS) stars gives a greater chance to investigate the interiors of stars. We analyzed seven HADS stars showing peak-to-peak amplitudes of more than 0.3 mag that were newly observed by TESS. We obtained that TIC374753270, TIC710783, and TIC187386415 pulsate in fundamental radial mode, also, TIC130474019 and TIC160120432 show double radial modes. On the other hand, TIC148357344 and TIC 278119167 demonstrate triple-mode behavior. Our analysis shows that these seven stars are close to the red edge of the (inside) instability strip in the Hertzsprung Russell diagram. The fundamental mode of these seven targets follows the period-luminosity (PL) relation for Delta Sct stars. However, TIC278119167 deviates slightly from the fundamental PL relation. The double-mode and triple-mode HADS stars (TIC130474019, TIC160120432, TIC148357344, and TIC278119167) are in agreement with the period ratio ranges (fundamental to first and second overtones). Using the information of 176 HADS stars (Netzel and Smolec), we find a scaling relation ([Fe/H] ~log((M**7.95+-0.15)(L**-1.83+-0.11)(P0**0.79+-0.14)(Teff**0.047+-0.02))) between the metallicity ([Fe/H]) and mass (M), luminosity (L), effective temperature (Teff), and the fundamental period (P0). We estimate the metallicity of the seven newly identified HADS stars ranging from -0.62 to 0.37 dex.

Purpose: Apparent diffusion coefficient (ADC) maps derived from diffusion weighted (DWI) MRI provides functional measurements about the water molecules in tissues. However, DWI is time consuming and very susceptible to image artifacts, leading to inaccurate ADC measurements. This study aims to develop a deep learning framework to synthesize ADC maps from multi-parametric MR images. Methods: We proposed the multiparametric residual vision transformer model (MPR-ViT) that leverages the long-range context of ViT layers along with the precision of convolutional operators. Residual blocks throughout the network significantly increasing the representational power of the model. The MPR-ViT model was applied to T1w and T2- fluid attenuated inversion recovery images of 501 glioma cases from a publicly available dataset including preprocessed ADC maps. Selected patients were divided into training (N=400), validation (N=50) and test (N=51) sets, respectively. Using the preprocessed ADC maps as ground truth, model performance was evaluated and compared against the Vision Convolutional Transformer (VCT) and residual vision transformer (ResViT) models. Results: The results are as follows using T1w + T2-FLAIR MRI as inputs: MPR-ViT - PSNR: 31.0 +/- 2.1, MSE: 0.009 +/- 0.0005, SSIM: 0.950 +/- 0.015. In addition, ablation studies showed the relative impact on performance of each input sequence. Both qualitative and quantitative results indicate that the proposed MR- ViT model performs favorably against the ground truth data. Conclusion: We show that high-quality ADC maps can be synthesized from structural MRI using a MPR- VCT model. Our predicted images show better conformality to the ground truth volume than ResViT and VCT predictions. These high-quality synthetic ADC maps would be particularly useful for disease diagnosis and intervention, especially when ADC maps have artifacts or are unavailable., Comment: arXiv admin note: text overlap with arXiv:2311.15044