Safe and Balanced: A Framework for Constrained Multi-Objective Reinforcement Learning.

In numerous reinforcement learning (RL) problems involving safety-critical systems, a key challenge lies in balancing multiple objectives while simultaneously meeting all stringent safety constraints. To tackle this issue, we propose a primal-based framework that orchestrates policy optimization between multi-objective learning and constraint adherence. Our method employs a novel natural policy gradient manipulation method to optimize multiple RL objectives and overcome conflicting gradients between different tasks, since the simple weighted average gradient direction may not be beneficial for specific tasks' performance due to misaligned gradients of different task objectives.

Prognostic Significance of the Detection of Human Papilloma Virus L1 Protein in Smears of Cervical Intraepithelial Neoplasia Grade 3 in Pregnant Women.

Background/aim: Cervical intraepithelial neoplasia (CIN) III/high-grade squamous lesions (HSIL) remains a significant challenge during pregnancy. Current data on the course of disease are contradictory, with cases of progression to cervical cancer (CC) during pregnancy being observed. Evidence suggests that the expression of L1 capsid protein is associated with a favorable prognosis in non-pregnant women.

Assessing Genetic Diversity and Population Structure of Western Honey Bees in the Czech Republic Using 22 Microsatellite Loci.

To date, no study has been conducted to investigate the diversity in honeybee populations of in the Czech Republic. Between 2022 and 2023, worker bees were collected from colonies distributed throughout the Czech Republic in 77 districts, and their genetic differences were examined using 22 microsatellite loci. The samples were obtained from hives ( = 3647) and through the process of capture on flowers ( = 553).

Clinical Outcome After Surgical Treatment of Traumatic Peroneal Nerve Injury: An Analysis of Risk Factors After Different Surgical Approaches.

Background: This study aims to analyze potential risk factors that may influence the clinical outcomes following surgical treatment of traumatic peroneal nerve lesions.

Methods: We conducted a retrospective analysis of patients with traumatic peroneal nerve injuries treated with decompression, split repair, or nerve grafting between 2010 and 2020. Motor function and potential risk factors were evaluated.

PD-L1-Expression in primary and recurrent vulvar squamous cell cancer.

Background: Squamous cell vulvar carcinoma is a rare malignant disease of women. In higher tumor stages survival rates are poor. Therapy options are limited.

The Role of Transcranial Ultrasound Imaging in Intensive Care Treatment of Decompressive Hemicraniectomy Patients: A Retrospective Single-Center Analysis.

: Post-hemicraniectomy patients often need extended intensive care treatment. While computed tomography (CT) is considered the gold standard for regular imaging, its frequent use could be linked to adverse clinical outcomes. This study aimed to assess bedside transcranial ultrasound (TUS) to capture intracranial anatomical structures and pathologies.

Modeling Grid Cell Distortions with a Grid Cell Calibration Mechanism.

The medial entorhinal cortex of rodents is known to contain grid cells that exhibit precise periodic firing patterns based on the animal's position, resulting in a distinct hexagonal pattern in space. These cells have been extensively studied due to their potential to unveil the navigational computations that occur within the mammalian brain and interesting phenomena such as so-called grid cell distortions have been observed. Previous neuronal models of grid cells assumed their firing fields were independent of environmental boundaries.

Light harvesting FIT DNA hybridization probes for brightness-enhanced RNA detection.

Fluorogenic hybridization probes are essential tools in modern molecular biology techniques. They allow detection of specific nucleic acid molecules without the need to separate target-bound from unbound probes. To enable detection of targets at low concentration, fluorogenic probes should have high brightness.

Catch the wind: Optimizing wind turbine power generation by addressing wind veer effects.

Wind direction variability with height, known as "wind veer," results in power losses for wind turbines (WTs) that rely on single-point wind measurements at the turbine nacelles. To address this challenge, we introduce a yaw control strategy designed to optimize turbine alignment by adjusting the yaw angle based on specific wind veer conditions, thereby boosting power generation efficiency. This strategy integrates modest yaw offset angles into the existing turbine control systems via a yaw-bias-look-up table, which correlates the adjustments with wind speed, and wind veer data.

Ultrastructural Perspectives on the Biology and Taphonomy of Tonian Microfossils From the Draken Formation, Spitsbergen.

Silicified peritidal carbonates of the Tonian Draken Formation, Spitsbergen, contain highly diverse and well-preserved microfossil assemblages dominated by filamentous microbial mats, but also including diverse benthic and/or allochthonous (possibly planktonic) microorganisms. Here, we characterize eight morphospecies in focused ion beam (FIB) ultrathin sections using transmission electron microscopy (TEM) and X-ray absorption near-edge structure (XANES) spectromicroscopy. Raman and XANES spectroscopies show the highly aromatic molecular structure of preserved organic matter.

Effect of a giant meteorite impact on Paleoarchean surface environments and life.

Large meteorite impacts must have strongly affected the habitability of the early Earth. Rocks of the Archean Eon record at least 16 major impact events, involving bolides larger than 10 km in diameter. These impacts probably had severe, albeit temporary, consequences for surface environments.

A Review of Safe Reinforcement Learning: Methods, Theories, and Applications.

Reinforcement Learning (RL) has achieved tremendous success in many complex decision-making tasks. However, safety concerns are raised during deploying RL in real-world applications, leading to a growing demand for safe RL algorithms, such as in autonomous driving and robotics scenarios. While safe control has a long history, the study of safe RL algorithms is still in the early stages.

Early Pennsylvanian Lagerstätte reveals a diverse ecosystem on a subhumid, alluvial fan.

Much of what we know about terrestrial life during the Carboniferous Period comes from Middle Pennsylvanian (~315-307 Mya) Coal Measures deposited in low-lying wetland environments. We know relatively little about terrestrial ecosystems from the Early Pennsylvanian, which was a critical interval for the diversification of insects, arachnids, tetrapods, and seed plants. Here we report a diverse Early Pennsylvanian trace and body fossil Lagerstätte (~320-318 Mya) from the Wamsutta Formation of eastern North America, distinct from coal-bearing deposits, preserved in clastic substrates within basin margin conglomerates.

Transformation Decoupling Strategy Based on Screw Theory for Deterministic Point Cloud Registration With Gravity Prior.

Point cloud registration is challenging in the presence of heavy outlier correspondences. This paper focuses on addressing the robust correspondence-based registration problem with gravity prior that often arises in practice. The gravity directions are typically obtained by inertial measurement units (IMUs) and can reduce the degree of freedom (DOF) of rotation from 3 to 1.

Robotic scrub nurse to anticipate surgical instruments based on real-time laparoscopic video analysis.

Background: Machine learning and robotics technologies are increasingly being used in the healthcare domain to improve the quality and efficiency of surgeries and to address challenges such as staff shortages. Robotic scrub nurses in particular offer great potential to address staff shortages by assuming nursing tasks such as the handover of surgical instruments.

Methods: We introduce a robotic scrub nurse system designed to enhance the quality of surgeries and efficiency of surgical workflows by predicting and delivering the required surgical instruments based on real-time laparoscopic video analysis.

Haplotype Diversity in mtDNA of Honeybee in the Czech Republic Confirms Complete Replacement of Autochthonous Population with the C Lineage.

The study aimed to analyze the genetic diversity in the Czech population of using mitochondrial DNA markers, intergenic region and gene. A total of 308 samples of bees were collected from the entire Czech Republic (from colonies and flowers in 13 different regions). Following sequencing, several polymorphisms and haplotypes were identified.

Unveiling Challenging Microbial Fossil Biosignatures from Rio Tinto with Micro-to-Nanoscale Chemical and Ultrastructural Imaging.

Understanding the nature and preservation of microbial traces in extreme environments is crucial for reconstructing Earth's early biosphere and for the search for life on other planets or moons. At Rio Tinto, southwestern Spain, ferric oxide and sulfate deposits similar to those discovered at Meridiani Planum, Mars, entomb a diversity of fossilized organisms, despite chemical conditions commonly thought to be challenging for life and fossil preservation. Investigating this unique fossil microbiota can elucidate ancient extremophile communities and the preservation of biosignatures in acidic environments on Earth and, potentially, Mars.

Towards multimodal graph neural networks for surgical instrument anticipation.

Purpose: Decision support systems and context-aware assistance in the operating room have emerged as the key clinical applications supporting surgeons in their daily work and are generally based on single modalities. The model- and knowledge-based integration of multimodal data as a basis for decision support systems that can dynamically adapt to the surgical workflow has not yet been established. Therefore, we propose a knowledge-enhanced method for fusing multimodal data for anticipation tasks.

Revitalizing Convolutional Network for Image Restoration.

Image restoration aims to reconstruct a high-quality image from its corrupted version, playing essential roles in many scenarios. Recent years have witnessed a paradigm shift in image restoration from convolutional neural networks (CNNs) to Transformer-based models due to their powerful ability to model long-range pixel interactions. In this paper, we explore the potential of CNNs for image restoration and show that the proposed simple convolutional network architecture, termed ConvIR, can perform on par with or better than the Transformer counterparts.

Grid codes vs. multi-scale, multi-field place codes for space.

Introduction: Recent work on bats flying over long distances has revealed that single hippocampal cells have multiple place fields of different sizes. At the network level, a multi-scale, multi-field place cell code outperforms classical single-scale, single-field place codes, yet the performance boundaries of such a code remain an open question. In particular, it is unknown how general multi-field codes compare to a highly regular grid code, in which cells form distinct modules with different scales.

Magnetic Soft Microrobot Design for Cell Grasping and Transportation.

Manipulating cells at a small scale is widely acknowledged as a complex and challenging task, especially when it comes to cell grasping and transportation. Various precise methods have been developed to remotely control the movement of microrobots. However, the manipulation of micro-objects necessitates the use of end-effectors.