Explore the vast range of titles available.

With rapid urbanization and industrialization, rural rivers in China are facing deterioration in water quality and ecosystem health. Microorganisms living in river sediments are involved in biogeochemical processes, mineralization, and degradation of pollutants. Understanding bacterial community distribution in rural rivers could help evaluate the response of river ecosystems to environmental pollution and understand the river self-purification mechanism. In this study, the relationship between characteristics of sediment microbial communities and the surrounding environmental factors in a typical rural river was analyzed using 16S rRNA gene sequencing technology. The results showed that the dominant bacterial groups in the river sediment were Proteobacteria , Actinobacteria , Chloroflexi , Acidobacteria , Bacteroidetes , and Firmicutes , accounting for 83.61% of the total microbial load. Different areas have different sources of pollution which give rise to specific dominant bacteria. The upstream part of the river flows through an agricultural cultivation area where the dominant bacteria were norank_f_Gemmatimonadaceae , Haliangium , and Pseudolabrys , possessing obvious nitrogen- and phosphorus-metabolizing activities. The midstream section flows through an urban area where the dominant bacteria were Marmoricola , Nocardioides , Gaiella , Sphingomonas , norank_f_67-14 , Subgroup_10 , Agromyces , and Lysobacter , with strong metabolizing activity for toxic pollutants. The dominant bacteria in the downstream part were Clostridium_sensu_stricto_1 , norank_f__Bacteroidetes_vadinHA17 , Candidatus_Competibacter , and Methylocystis . Redundancy analysis and correlation heatmap analysis showed that environmental factors: ammonia nitrogen (NH 4 + –N) and total nitrogen (TN) in the sediment, and pH, temperature, TN, electrical conductivity (EC), and total dissolved solids (TDS) in the water, significantly affected the bacterial community in the sediment. The PICRUSt2 functional prediction analysis identified that the main function of bacteria in the sediment was metabolism (77.3%), specifically carbohydrate, amino acid, and energy metabolism. These activities are important for degrading organic matter and removing pollutants from the sediments. The study revealed the influence of organic pollutants derived from human activities on the bacterial community composition in the river sediments. It gave a new insight into the relationship between environmental factors and bacterial community distribution in rural watershed ecosystems, providing a theoretical basis for self-purification and bioremediation of rural rivers.

To assess the health of river ecosystems, it is essential to quantify the ecological risk of heavy metals in river sediments and the structure of microbial communities. As important tributaries of the Tuo River in the upper reaches of the Yangtze River, the Mianyuan River and the Shiting River, are closely related to the economic development and human daily life in the region. This study assessed the ecological risks of heavy-metal-polluted river sediments, the heavy-metal-driven bacterial communities were revealed, and the relationships between the ecological risks and the identical bacterial communities were discussed. The Cd content was significantly greater than the environmental background value, leading to a serious pollution and very high ecological risk at the confluence of the two rivers and the upper reaches of the Mianyuan River. Microbial community analysis showed that Rhodobacter, Nocardioides, Sphingomonas, and Pseudarthrobacter were the dominant bacterial genera in the sediments of the Shiting River. However, the dominant bacterial genera in the Mianyuan River were Kouleothrix, Dechloromonas, Gaiella, Pedomicrobium, and Hyphomicrobium. Mantel test results showed (r = 0.5977, P = 0.005) that the Cd, As, Zn, Pb, Cr, and Cu were important factors that influenced differences in the distribution of sediment bacterial communities Mianyuan and Shiting rivers. A correlation heatmap showed that heavy metals were negatively correlated for most bacterial communities, but some bacterial communities were tolerant and showed a positive correlation. Overall, the microbial structure of the river sediments showed a diverse spatial distribution due to the influence of heavy metals. The results will improve the understanding of rivers contaminated by heavy metals and provide theoretical support for conservation and in situ ecological restoration of river ecosystems.

In this paper, we study the robustness of classical deep hedging strategies under distributional shifts by leveraging the concept of adversarial attacks. We first demonstrate that standard deep hedging models are highly vulnerable to small perturbations in the input distribution, resulting in significant performance degradation. Motivated by this, we propose an adversarial training framework tailored to increase the robustness of deep hedging strategies. Our approach extends pointwise adversarial attacks to the distributional setting and introduces a computationally tractable reformulation of the adversarial optimization problem over a Wasserstein ball. This enables the efficient training of hedging strategies that are resilient to distributional perturbations. Through extensive numerical experiments, we show that adversarially trained deep hedging strategies consistently outperform their classical counterparts in terms of out-of-sample performance and resilience to model misspecification. Additional results indicate that the robust strategies maintain reliable performance on real market data and remain effective during periods of market change. Our findings establish a practical and effective framework for robust deep hedging under realistic market uncertainties.

Adversarial training is a standard defense against malicious input perturbations in security-critical machine-learning systems. Its main burden is structural: before every parameter update, the current model must first be attacked to find a new adversarial perturbation, making training increasingly expensive and hard to sustain at large-model scale. Here we give an end-to-end quantum procedure for projected-gradient robust training under local stability and sparsity assumptions. The key step is to reformulate the coupled attacker--learner dynamics as a high-dimensional sparse linear system whose terminal block yields the final network-parameter state. In this formulation, the dominant query cost scales linearly with training time steps, up to logarithmic factors, and polylogarithmically with model size, while the full gate complexity records separate input-preparation and sparse-access overheads. This places core computational tasks for AI security on a concrete quantum footing and identifies a regime in which robust-training overhead can be reduced.

Design of adversarial attacks for deep neural networks, as well as methods of adversarial training against them, are subject of intense research. In this paper, we propose methods to train against distributional attack threats, extending the TRADES method used for pointwise attacks. Our approach leverages recent contributions and relies on sensitivity analysis for Wasserstein distributionally robust optimization problems. We introduce an efficient fine-tuning method which can be deployed on a previously trained model. We test our methods on a range of pre-trained models on RobustBench. These experimental results demonstrate the additional training enhances Wasserstein distributional robustness, while maintaining original levels of pointwise robustness, even for already very successful networks. The improvements are less marked for models pre-trained using huge synthetic datasets of 20-100M images. However, remarkably, sometimes our methods are still able to improve their performance even when trained using only the original training dataset (50k images).

Deep neural networks are known to be vulnerable to adversarial attacks (AA). For an image recognition task, this means that a small perturbation of the original can result in the image being misclassified. Design of such attacks as well as methods of adversarial training against them are subject of intense research. We re-cast the problem using techniques of Wasserstein distributionally robust optimization (DRO) and obtain novel contributions leveraging recent insights from DRO sensitivity analysis. We consider a set of distributional threat models. Unlike the traditional pointwise attacks, which assume a uniform bound on perturbation of each input data point, distributional threat models allow attackers to perturb inputs in a non-uniform way. We link these more general attacks with questions of out-of-sample performance and Knightian uncertainty. To evaluate the distributional robustness of neural networks, we propose a first-order AA algorithm and its multi-step version. Our attack algorithms include Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) as special cases. Furthermore, we provide a new asymptotic estimate of the adversarial accuracy against distributional threat models. The bound is fast to compute and first-order accurate, offering new insights even for the pointwise AA. It also naturally yields out-of-sample performance guarantees. We conduct numerical experiments on the CIFAR-10 dataset using DNNs on RobustBench to illustrate our theoretical results. Our code is available at https://github.com/JanObloj/W-DRO-Adversarial-Methods.

Predicting tissue expression profiles from peripheral 'surrogate' samples, especially blood transcriptome, has become an effective alternative when invasive procedures are not ideal. However, existing approaches ignore tissue-shared intrinsic relevance, inevitably limiting predictive performance. Here, we propose a unified deep learning-based multi-task learning framework, Multi-tissue Transcriptome Mapping (MTM), enabling the prediction of individualized expression profiles from any available tissue of an individual. By jointly leveraging individualized cross-tissue information through multi-task learning, MTM achieves superior sample-level and gene-level performance. With the high prediction accuracy and the ability to preserve individualized biological variations, MTM could facilitate both fundamental and clinical biomedical research. Competing Interest Statement The authors have declared no competing interest. Footnotes * Title and abstract changed; keywords added.

With the rapid development of object detection technology for unmanned aerial vehicles (UAVs), it is convenient to collect data from UAV aerial photographs. They have a wide range of applications in several fields, such as monitoring, geological exploration, precision agriculture, and disaster early warning. In recent years, many methods based on artificial intelligence have been proposed for UAV object detection, and deep learning is a key area in this field. Significant progress has been achieved in the area of deep-learning-based UAV object detection. Thus, this paper presents a review of recent research on deep-learning-based UAV object detection. This survey provides an overview of the development of UAVs and summarizes the deep-learning-based methods in object detection for UAVs. In addition, the key issues in UAV object detection are analyzed, such as small object detection, object detection under complex backgrounds, object rotation, scale change, and category imbalance problems. Then, some representative solutions based on deep learning for these issues are summarized. Finally, future research directions in the field of UAV object detection are discussed.

Decision trees are popular classification models, providing high accuracy and intuitive explanations. However, as the tree size grows the model interpretability deteriorates. Traditional tree-induction algorithms, such as C4.5 and CART, rely on impurity-reduction functions that promote the discriminative power of each split. Thus, although these traditional methods are accurate in practice, there has been no theoretical guarantee that they will produce small trees. In this paper, we justify the use of a general family of impurity functions, including the popular functions of entropy and Gini-index, in scenarios where small trees are desirable, by showing that a simple enhancement can equip them with complexity guarantees. We consider a general setting, where objects to be classified are drawn from an arbitrary probability distribution, classification can be binary or multi-class, and splitting tests are associated with non-uniform costs. As a measure of tree complexity, we adopt the expected cost to classify an object drawn from the input distribution, which, in the uniform-cost case, is the expected number of tests. We propose a tree-induction algorithm that gives a logarithmic approximation guarantee on the tree complexity. This approximation factor is tight up to a constant factor under mild assumptions. The algorithm recursively selects a test that maximizes a greedy criterion defined as a weighted sum of three components. The first two components encourage the selection of tests that improve the balance and the cost-efficiency of the tree, respectively, while the third impurity-reduction component encourages the selection of more discriminative tests. As shown in our empirical evaluation, compared to the original heuristics, the enhanced algorithms strike an excellent balance between predictive accuracy and tree complexity.

Artificially improving traits of cultivated alfalfa ( Medicago sativa L.), one of the most important forage crops, is challenging due to the lack of a reference genome and an efficient genome editing protocol, which mainly result from its autotetraploidy and self-incompatibility. Here, we generate an allele-aware chromosome-level genome assembly for the cultivated alfalfa consisting of 32 allelic chromosomes by integrating high-fidelity single-molecule sequencing and Hi-C data. We further establish an efficient CRISPR/Cas9-based genome editing protocol on the basis of this genome assembly and precisely introduce tetra-allelic mutations into null mutants that display obvious phenotype changes. The mutated alleles and phenotypes of null mutants can be stably inherited in generations in a transgene-free manner by cross pollination, which may help in bypassing the debate about transgenic plants. The presented genome and CRISPR/Cas9-based transgene-free genome editing protocol provide key foundations for accelerating research and molecular breeding of this important forage crop. Alfalfa is an important forage crop, but genetic improvement is challenging due to the lack of a reference genome and an efficient genome editing protocol. Here, the authors report the chromosome-level assembly of the autotetraploid genome and a CRISPR/Cas9-based transgene-free genome editing protocol.