Search

Coordinating multiple agents with diverse tasks and changing goals without interference is a challenge. Multiagent reinforcement learning (MARL) aims to develop effective communication and joint policies using group learning. Some of the previous approaches required each agent to maintain a set of networks independently, resulting in no consideration of interactions. Joint communication work causes agents receiving information unrelated to their own tasks. Currently, agents with different task divisions are often grouped by action tendency, but this can lead to poor dynamic grouping. This article presents a two-phase solution for multiple agents, addressing these issues. The first phase develops heterogeneous agent communication joint policies using a group communication MARL framework (GCMA). The framework employs a periodic grouping strategy, reducing exploration and communication redundancy by dynamically assigning agent group hidden features through hypernetwork and graph communication. The scheme efficiently utilizes resources for adapting to multiple similar tasks. In the second phase, each agent's policy network is distilled into a generalized simple network, adapting to similar tasks with varying quantities and sizes. GCMA is tested in complex environments, such as StarCraft II and unmanned aerial vehicle (UAV) take-off, showing its well-performing for large-scale, coordinated tasks. It shows GCMA's effectiveness for solid generalization in multitask tests with simulated pedestrians.

Protein function prediction is essential for disease treatment and drug development; yet, traditional biological experimental methods are less efficient in annotating protein function, and existing automated methods fail to fully leverage protein multisource data. Here, we present MSF-PFP, a computational framework that fuses multisource data features to predict protein function with high accuracy. Our framework designs specific models for feature extraction based on the characteristics of various data sources, including a global-local-individual strategy for local location features. MSF-PFP then integrates extracted features through a multisource feature fusion model, ultimately categorizing protein functions. Experimental results demonstrate that MSF-PFP outperforms eight state-of-the-art models, achieving FMaxscores of 0.542, 0.675, and 0.624 for the biological process (BP), molecular function (MF), and cellular component (CC), respectively. The source code and data set for MSF-PFP are available at https://swanhub.co/TianGua/MSF-PFP, facilitating further exploration and validation of the proposed framework. This study highlights the potential of multisource data fusion in enhancing protein function prediction, contributing to improved disease therapy and medication discovery strategies.

The outbreak of COVID-19 has caused a serious impact on the traditional logistics industry. Considering that the truck-drone collaborative delivery system can both reduce the risk of COVID-19 propagation and deliver supplies in a cost effective and timely manner, this paper introduces the Multiple visits Travelling Salesman Problem with Multiple Heterogeneous Drones (MTSP-MHD). The model allows a truck to carry a fleet of heterogeneous multi-visit drones for cooperative deliveries, where the drones are capable of delivering to multiple customers on a single route and the flight is restricted by energy consumption and payload constraints. To solve MTSP-MHD, we develop an approach that combines K-Means++ clustering, Nearest neighbor search and Greedy strategies (KNG) to construct feasible solutions. Meanwhile, an Improved Artificial Bee Colony algorithm combining Metropolis acceptance criterion of Simulated Annealing, Tabu list of Tabu Search, and Elite selection strategies (IABC-MTE) is proposed to enhance the quality of solutions. Particularly, three problem-specific neighborhood operators are adopted to search for new solutions. The massive experimental results indicate that IABC-MTE achieves significant improvements over other competitors, with average objective value reductions ranging from 1.81% to 29.16% and standard deviations reduced by 0.04 to 26.44. Finally, the influencing factors of the drone fleet, the performance of different drone fleets and delivery modes are evaluated in detail.

A large number of experimental studies have shown that circRNAs can act as molecular sponges of microRNAs, interacting with miRNAs to regulate gene expression levels, thereby affecting the development of human diseases. Exploring the potential associations between circRNAs and miRNAs can help understand complex disease mechanisms. Considering that biological experiments are time-consuming and labor-intensive, this study proposes a computational model using a graph neural network and singular value decomposition (CMASG) for circRNA-miRNA association prediction. Specifically, graph neural networks are used to learn nonlinear feature representations of nodes, followed by matrix factorization algorithms to learn linear feature representations of nodes, and then combined feature representations learned from different perspectives. Finally, the lightGBM algorithm was used for circRNA-miRNA association prediction. The proposed CMASG model achieved an AUC value of 0.8804. The experimental results demonstrate the superiority and effectiveness of the CMASG model in predicting circRNA-miRNA association tasks.

Based on the depth feature of facial micro-expression recognition, such as by using Convolutional Neural Network, the method of classifying facial micro-expression recognition has been gradually improved. Compared with the traditional feature extraction method, the improved technique can more easily realize real-time application. To perfect the details and extract the fine features of micro-expressions, a new algorithm named MACNN (Multi-scale convolutional neural network model), which combines the kernels of atrous convolutions and automatic face correction, is proposed to improve the feature extraction process of the CNN network. Residual blocks are introduced to solve the gradient disappearance problem and to accelerate convergence during the model training. The model is trained and tested on the micro-expression public data sets CASME and CASME II through real-time detection in real-time application of automatic face correction. The robustness of the model is improved by comparing the loss function schemes. The accuracy of the method is 73%, and the real-time detection occurs at a frame rate of 60 FPS. This method can effectively improve the accuracy of micro-expression detection, meet the real-time requirements, and demonstrate satisfactory robustness and generalizability.

Noncoding RNA(ncRNA) is closely related to drug resistance. Identifying the association between ncRNA and drug resistance is of great significance for drug development. Methods based on biological experiments are often time-consuming and small-scale. Therefore, developing computational methods to distinguish the association between ncRNA and drug resistance is urgent. We develop a computational framework called GSLRDA to predict the association between ncRNA and drug resistance in this work. First, the known ncRNA–drug resistance associations are modeled as a bipartite graph of ncRNA and drug. Then, GSLRDA uses the light graph convolutional network (lightGCN) to learn the vector representation of ncRNA and drug from the ncRNA–drug bipartite graph. In addition, GSLRDA uses different data augmentation methods to generate different views for ncRNA and drug nodes and performs self-supervised learning, further improving the quality of learned ncRNA and drug vector representations through contrastive learning between nodes. Finally, GSLRDA uses the inner product to predict the association between ncRNA and drug resistance. To the best of our knowledge, GSLRDA is the first to apply self-supervised learning in association prediction tasks in the field of bioinformatics. The experimental results show that GSLRDA takes an AUC value of 0.9101, higher than the other eight state-of-the-art models. In addition, case studies including two drugs further illustrate the effectiveness of GSLRDA in predicting the association between ncRNA and drug resistance. The code and data sets of GSLRDA are available at https://github.com/JJZ-code/GSLRDA.

A vehicle make and model recognition (VMMR) system is a common requirement in the field of intelligent transportation systems (ITS). However, it is a challenging task because of the subtle differences between vehicle categories. In this paper, we propose a hierarchical scheme for VMMR. Specifically, the scheme consists of (1) a feature extraction framework called weighted mask hierarchical bilinear pooling (WMHBP) based on hierarchical bilinear pooling (HBP) which weakens the influence of invalid background regions by generating a weighted mask while extracting features from discriminative regions to form a more robust feature descriptor; (2) a hierarchical loss function that can learn the appearance differences between vehicle brands, and enhance vehicle recognition accuracy; (3) collection of vehicle images from the Internet and classification of images with hierarchical labels to augment data for solving the problem of insufficient data and low picture resolution and improving the model’s generalization ability and robustness. We evaluate the proposed framework for accuracy and real-time performance and the experiment results indicate a recognition accuracy of 95.1% and an FPS (frames per second) of 107 for the framework for the Stanford Cars public dataset, which demonstrates the superiority of the method and its availability for ITS.

Visible and near-infrared spectroscopy can be a powerful tool to rapidly quantify various soil characteristics. Recently, wavelet analysis has been proven to be efficient in many fields, including signal processing and digital image analysis. It can be used to smooth signals and to reduce large data sets to parsimonious representations. The objective of this study was to capture the absorption features of soil organic carbon (SOC) and total nitrogen (TN) by applying wavelet analysis to reflectance spectra. In addition, the correlation between SOC, TN contents, and soil spectra reflectance and wavelet coefficient were analyzed and compared. The results show that the maximum correlation coefficient between SOC, TN, and wavelet coefficient was more than 0.96 compared with the relationship between SOC, TN, and spectral reflectance (r= −0.79 for SOC, r= −0.40 for TN), especially for TN (the maximum negative correlation coefficient, r= −0.964). For SOC TN and SOCTN, because SOC contents accounted for a larger proportion of soil composition than TN, their spectral features were mainly affected by SOC in soil samples. In addition, wavelet analysis also enhanced the features of SOC TN and SOCTN obviously. These results suggest that wavelet analysis is a better method for capturing the absorption features of soil composition from hyperspectral data. Continuous wavelet transform was confirmed to be an essential and efficient tool used to locate the wavenumber where SOC and TN could be predicted with minimal interference from other components. The results also implied that continuous wavelet transform can be used in discriminating soil composition.

Most existing recommendation methods focus on the improvement of recommender accuracy while ignoring the influence of recommended explanation. Recommender explainability is an efficient way to help consumers make much more suitable decisions and enhance their acceptance and trustfulness in recommender systems (RSs). Incorporating a knowledge graph (KG) into RS is a promising way to improve recommender results while enhancing the strength of explanation. In this paper, an interactive rules-guided recommender (IR-Rec) framework based on KG is proposed. First, an existing KG is enriched by introducing it to facts about e-commerce. Then, a number of paths are extracted from the enhanced KG for user-item interactions that are able to ascertain the underlying reason for recommendations according to the semantic strength of the KG. These paths are summarized into some common behavior rules that have the ability to explain the underlying motivations of users. According to the characteristics of users, items, and rules, different neural networks are designed, such as a graph convolutional network, to learn more accurate embeddings. Furthermore, recommendations that meet users' personalized interests are made by combining the public behavior rules and their individual features. Extensive experiments are carried out on four Amazon datasets for top- K recommendation. All the results show that the proposed method performs better than other respective baselines and demonstrate the effectiveness of rule extraction for making recommendations. [ABSTRACT FROM AUTHOR]