Explore the vast range of titles available.

Various strategies for active learning have been proposed in the machine learning literature. In uncertainty sampling, which is among the most popular approaches, the active learner sequentially queries the label of those instances for which its current prediction is maximally uncertain. The predictions as well as the measures used to quantify the degree of uncertainty, such as entropy, are traditionally of a probabilistic nature. Yet, alternative approaches to capturing uncertainty in machine learning, alongside with corresponding uncertainty measures, have been proposed in recent years. In particular, some of these measures seek to distinguish different sources and to separate different types of uncertainty, such as the reducible (epistemic) and the irreducible (aleatoric) part of the total uncertainty in a prediction. The goal of this paper is to elaborate on the usefulness of such measures for uncertainty sampling, and to compare their performance in active learning. To this end, we instantiate uncertainty sampling with different measures, analyze the properties of the sampling strategies thus obtained, and compare them in an experimental study.

Background Single-cell RNA sequencing (scRNA-seq) enables the identification of cell types within complex biological systems, yet accurately classifying both known and novel cell types remains a significant challenge. Supervised learning methods perform well when all cell types are labeled in the training data, but struggle with unseen cell types, while rejection-based approaches can mitigate misclassification but fail to leverage unlabeled data for learning. Deep learning-based methods, such as MARS, offer promising solutions but often suffer from poor generalization to novel cell populations. Results We propose Semi-supervised learning for Robust Novel Cell-type identification (SRNC), a novel semi-supervised framework that enhances classification accuracy while effectively identifying unknown cell types. By integrating self-supervised feature learning with semi-supervised classification, SRNC leverages both labeled and unlabeled data to improve generalization. Evaluated across six benchmark scRNA-seq datasets, SRNC consistently outperforms state-of-the-art methods, achieving higher ARI, F1-score, and precision than both the rejection-based approach and deep-learning-based MARS. Moreover, SRNC demonstrates robustness across datasets from different laboratories and excels in imbalanced classification scenarios, accurately identifying rare cell populations that other methods often misclassify. Conclusions Our results demonstrate that SRNC is a powerful and adaptable tool for cell-type classification in scRNA-seq analysis. By leveraging semi-supervised learning, SRNC effectively identifies both known and novel cell types, surpassing competing methods in multiple performance metrics. Its ability to generalize across datasets and handle class imbalances makes it a valuable approach for discovering new cell types, advancing precision medicine, and improving our understanding of cellular heterogeneity.

In contrast to conventional (single-label) classification, the setting of multilabel classification (MLC) allows an instance to belong to several classes simultaneously. Thus, instead of selecting a single class label, predictions take the form of a subset of all labels. In this paper, we study an extension of the setting of MLC, in which the learner is allowed to partially abstain from a prediction, that is, to deliver predictions on some but not necessarily all class labels. This option is useful in cases of uncertainty, where the learner does not feel confident enough on the entire label set. Adopting a decision-theoretic perspective, we propose a formal framework of MLC with partial abstention, which builds on two main building blocks: First, the extension of underlying MLC loss functions so as to accommodate abstention in a proper way, and second the problem of optimal prediction, that is, finding the Bayes-optimal prediction minimizing this generalized loss in expectation. It is well known that different (generalized) loss functions may have different risk-minimizing predictions, and finding the Bayes predictor typically comes down to solving a computationally complexity optimization problem. In the most general case, given a prediction of the (conditional) joint distribution of possible labelings, the minimizer of the expected loss needs to be found over a number of candidates which is exponential in the number of class labels. We elaborate on properties of risk minimizers for several commonly used (generalized) MLC loss functions, show them to have a specific structure, and leverage this structure to devise efficient methods for computing Bayes predictors. Experimentally, we show MLC with partial abstention to be effective in the sense of reducing loss when being allowed to abstain.

Chitosan (CH) shows great potential as an immunostimulatory feed additive in aquaculture. This study evaluates the effects of varying dietary CH levels on the growth, immunity, intestinal morphology, and antioxidant status of Nile tilapia (Oreochromis niloticus) reared in a biofloc system. Tilapia fingerlings (mean weight 13.54 ± 0.05 g) were fed diets supplemented with 0 (CH0), 5 (CH5), 10 (CH10), 20 (CH20), and 40 (CH40) mL·kg−1 of CH for 8 weeks. Parameters were assessed after 4 and 8 weeks. Their final weight was not affected by CH supplementation, but CH at 10 mL·kg−1 significantly improved weight gain (WG) and specific growth rate (SGR) compared to the control (p < 0.05) at 8 weeks. Skin mucus lysozyme and peroxidase activities were lower in the chitosan-treated groups at weeks 4 and 8. Intestinal villi length and width were enhanced by 10 and 20 mL·kg−1 CH compared to the control. However, 40 mL·kg−1 CH caused detrimental impacts on the villi and muscular layer. CH supplementation, especially 5–10 mL·kg−1, increased liver and intestinal expressions of interleukin 1 (IL-1), interleukin 8 (IL-8), LPS-binding protein (LBP), glutathione reductase (GSR), glutathione peroxidase (GPX), and glutathione S-transferase (GST-α) compared to the control group. Overall, dietary CH at 10 mL·kg−1 can effectively promote growth, intestinal morphology, innate immunity, and antioxidant capacity in Nile tilapia fingerlings reared in biofloc systems.

Previously, we reported that overexpression of AtRH17, an Arabidopsis DEAD-box RNA helicase gene, confers salt stress-tolerance via a pathway other than the well-known salt stress-responsive pathways. To decipher the salt stress-responsive pathway in AtRH17-overexpressing transgenic plants (OXs), we performed RNA-Sequencing and identified 397 differentially expressed genes between wild type (WT) and AtRH17 OXs. Among them, 286 genes were upregulated and 111 genes were downregulated in AtRH17 OXs relative to WT. Gene ontology annotation enrichment and KEGG pathway analysis showed that the 397 upregulated and downregulated genes are involved in various biological functions including secretion, signaling, detoxification, metabolic pathways, catabolic pathways, and biosynthesis of secondary metabolites as well as in stress responses. Genevestigator analysis of the upregulated genes showed that nine genes, namely, LEA4-5, GSTF6, DIN2/BGLU30, TSPO, GSTF7, LEA18, HAI1, ABR, and LTI30, were upregulated in Arabidopsis under salt, osmotic, and drought stress conditions. In particular, the expression levels of LEA4-5, TSPO, and ABR were higher in AtRH17 OXs than in WT under salt stress condition. Taken together, our results suggest that a high AtRH17 expression confers salt stress-tolerance through a novel salt stress-responsive pathway involving nine genes, other than the well-known ABA-dependent and ABA-independent pathways.

Quantum compilation is the process of converting a target unitary operation into a trainable unitary represented by a quantum circuit. It has a wide range of applications, including gate optimization, quantum-assisted compiling, quantum state preparation, and quantum dynamic simulation. Traditional quantum compilation usually optimizes circuits for a single target. However, many quantum systems require simultaneous optimization of multiple targets, such as thermal state preparation, time-dependent dynamic simulation, and others. To address this, we develop a multi-target quantum compilation algorithm to improve the performance and flexibility of simulating multiple quantum systems. Our benchmarks and case studies demonstrate the effectiveness of the algorithm, highlighting the importance of multi-target optimization in advancing quantum computing. This work lays the groundwork for further development and evaluation of multi-target quantum compilation algorithms.

Insect meal has shown promise as a potentially sustainable source of nutrients for aquafeeds, offering an alternative to expensive and ecologically undesirable ingredients, in the context of population explosion and climate change. Despite this promising outlook, its effects on fish growth and immune responses remain to be thoroughly investigated. Our scientific goal was to experimentally test responses to replacements of the fish meal with a protein source derived from black soldier fly larvae meal (BSFLM). Possible impacts on growth, immunological response, and the expression of selected immune-system related genes were evaluated in Koi carp (Cyprinus carpio var. koi) using a biofloc culture system. Three hundred fish (20.0 ± 0.2 g) were allocated into five groups: a control group receiving a basal diet containing 0 g kg−1 BSFLM and four experimental groups in which fish meal was replaced with 50, 100, 150, and 200 g kg−1 BSFLM for eight weeks. After 4 weeks of feeding, there were no statistically significant differences in specific growth rate (SGR), feed conversion ratio (FCR), and survival rate between fish fed BSFLM-enriched diets at 50, 100, 150 g kg−1 and a control (0 g kg−1 BSFLM) diet. However, fish fed 200 g kg−1 BSFLM showed significantly improved weight gain (WG) and SGR compared to the control after 4 weeks; this difference persisted through 8 weeks (p < 0.05). After eight weeks, there was a moderate to weak negative linear regression shown in FCR (r = 0.470) and SR (r = 0.384), respectively, with the BSFLM levels, but significant and highly correlated linear relationships were observed in WG (r = 0.917) and SGR (r = 0.912). Immunological response analysis showed slight changes in lysozyme and peroxidase levels by replacing fish meal with BSFLM, but these apparent differences were not significantly related to experimental diets. Interestingly, mRNA transcripts of immune-related genes (TNF-α, TGF-β, IL1, IL10, and hsp70) were upregulated in the groups receiving higher amounts of BSFLM, with statistically significant differences observed in certain comparisons. Our findings reveal that fish meal can be effectively replaced by BSFLM, and that this not only has a positive effect on immune-related gene expression in Koi carp, but also on growth rate, pointing to the future potential role of BSFLM as an alternative fish meal protein in aquafeed formulation.

Vatica diospyroides , an endemic species of the Dipterocarpaceae family, possesses notable medicinal properties. However, its application as an antibacterial agent is limited due to the insufficient investigations of its antibacterial activity from flower extracts. This study is aimed at exploring the antibacterial mechanisms of acetone extracts from the flowers of V. diospyroides against four bacterial strains using various methods, including the well‐disk diffusion assay, minimum inhibitory concentration (MIC) determination, minimum bactericidal concentration (MBC) assessment, flow cytometry, and scanning electron microscopy. The inhibition zones measured between 6.33 and 17.66 mm. Notably, the extract exhibited different MIC values, such as 250 μ g mL -1 for Bacillus subtilis and Escherichia coli , and only 62.5 μ g.mL -1 for Vibrio parahaemolyticus , demonstrating its effectiveness. MBC values ranged from 500 to over 1000 μ g mL -1 for Pseudomonas aeruginosa . Flow cytometric analysis revealed that the cellular responses to the extract were influenced by both the concentration of the extract and the duration of exposure, indicating a dose‐ and time‐dependent antibacterial effect. Additionally, scanning electron microscopy confirmed that the extract caused structural damage to the cells of both Gram‐positive and Gram‐negative bacteria. Overall, this study underscores the promising antibacterial potential of V. diospyroides flower extracts, which demonstrate significant efficacy against a variety of bacterial strains.

Highlights FCS improved koi carp growth, feed efficiency, and SGR. Antioxidant capacity enhanced ABTS scavenging, decreased MDA in FCS-fed fish. FCS increased Bacillus/Lactobacillus and reduced Aeromonas abundance.

Dealing with partially known or missing data is a common problem in machine learning. This work is interested in the problem of querying the true value of data to improve the quality of the learned model, when those data are only partially known. This study is in the line of active learning, since we consider that the precise value of some partial data can be queried to reduce the uncertainty in the learning process, yet can consider any kind of partial data (not only entirely missing one). We propose a querying strategy based on the concept of racing algorithms in which several models are competing. The idea is to identify the query that will help the most to quickly decide the winning model in the competition. After discussing and formalizing the general ideas of our approach, we study the particular case of decision trees in case of interval-valued features and set-valued labels. The experimental results indicate that, in comparison with other baselines, the proposed approach significantly outperforms simpler strategies in the case of partially specified features, while it achieves similar performances in the case of partially specified labels.