Learning Framework Research Articles

GCLmf: A Novel Molecular Graph Contrastive Learning Framework Based on Hard Negatives and Application in Toxicity Prediction.

In silico methods for prediction of chemical toxicity can decrease the cost and increase the efficiency in the early stage of drug discovery. However, due to low accessibility of sufficient and reliable toxicity data, constructing robust and accurate prediction models is challenging. Contrastive learning, a type of self-supervised learning, leverages large unlabeled data to obtain more expressive molecular representations, which can boost the prediction performance on downstream tasks. While molecular graph contrastive learning has gathered growing attentions, current models neglect the quality of negative data set. Here, we proposed a self-supervised pretraining deep learning framework named GCLmf. We first utilized molecular fragments that meet specific conditions as hard negative samples to boost the quality of the negative set and thus increase the difficulty of the proxy tasks during pre-training to learn informative representations. GCLmf has shown excellent predictive power on various molecular property benchmarks and demonstrates high performance in 33 toxicity tasks in comparison with multiple baselines. In addition, we further investigated the necessity of introducing hard negatives in model building and the impact of the proportion of hard negatives on the model.

Empowering Mental Health Monitoring Using a Macro-Micro Personalization Framework for Multimodal-Multitask Learning: Descriptive Study.

The field of mental health technology presently has significant gaps that need addressing, particularly in the domain of daily monitoring and personalized assessments. Current noninvasive devices such as wristbands and smartphones are capable of collecting a wide range of data, which has not yet been fully used for mental health monitoring. This study aims to introduce a novel dataset for personalized daily mental health monitoring and a new macro-micro framework. This framework is designed to use multimodal and multitask learning strategies for improved personalization and prediction of emotional states in individuals. Data were collected from 298 individuals using wristbands and smartphones, capturing physiological signals, speech data, and self-annotated emotional states. The proposed framework combines macro-level emotion transformer embeddings with micro-level personalization layers specific to each user. It also introduces a Dynamic Restrained Uncertainty Weighting method to effectively integrate various data types for a balanced representation of emotional states. Several fusion techniques, personalization strategies, and multitask learning approaches were explored. The proposed framework was evaluated using the concordance correlation coefficient, resulting in a score of 0.503. This result demonstrates the framework's efficacy in predicting emotional states. The study concludes that the proposed multimodal and multitask learning framework, which leverages transformer-based techniques and dynamic task weighting strategies, is superior for the personalized monitoring of mental health. The study indicates the potential of transforming daily mental health monitoring into a more personalized app, opening up new avenues for technology-based mental health interventions.

Data-driven Strategies for Affordable Housing: A Hybrid Genetic Algorithm-Machine Learning Optimization Model in the Melbourne Metropolitan Area

Abstract. The escalating rental crisis in Australia, rooted in demographic dynamics, a scarcity of lands suitable for development and the influence of interest rates and government subsidies, necessitates innovative solutions for housing affordability. One promising approach is to increase the supply of affordable rental housing. Nevertheless, these efforts must be guided by well-informed policies as they are required to cater to the diverse needs of the stakeholders involved. Accordingly, this study aims to prioritize suburbs in the Melbourne Metropolitan area for the construction of apartment/unit buildings, benefiting both builders/investors and renters. Leveraging a hybrid Genetic Algorithm-Machine Learning (GA-ML) framework, multi-objective optimization models are developed to rank 105 suburbs based on key parameters for both builders/investors and renters. The optimization process seeks to maximize economic outlook and rental yields for builders/investors while minimizing rents and commuting distances for renters, in addition to proximity to essential amenities. First, an initial optimization using GA based on six key parameters is performed considering a linear multi-objective function. Subsequently, ML-based objective functions are defined using Random Forest (RF) and extreme Gradient Boosting (XGBoost) models to refine the optimization model for suburb rankings. The evaluation reveals strong correlations between GA and GA-XGB rankings, suggesting the effectiveness of the GA-XGBoost model in prioritizing suburbs. Notably, suburbs consistently prioritized across all models include Brunswick, Coburg, Preston and Reservoir, highlighting their suitability for apartment/unit building construction. By directing attention to specific suburbs aligned with different stakeholders' needs and preferences, this study contributes to a more sustainable and equitable housing landscape.

Intelligent algorithmic framework for detection and mitigation of BeiDou spoofing attacks in vehicular ad hoc networks (VANETs)

This research tackles the critical challenge of BeiDou signal spoofing in vehicular ad-hoc networks and addresses significant risks to vehicular safety and traffic management stemming from increased reliance on accurate satellite navigation. The study proposes a novel hybrid machine learning framework that integrates Autoencoders and long short-term memory (LSTM) networks with an advanced cryptographic method, attribute-based encryption, to enhance the detection and mitigation of spoofing attacks. Our methodology leverages both real-time and synthetic navigational data in a comprehensive experimental setup that simulates various spoofing scenarios to test the resilience of the proposed system. The findings demonstrate a significant improvement in the accuracy of spoofing detection and the robustness of mitigation strategies by ensuring the integrity and reliability of navigational data. This investigation enhances the existing body of knowledge by demonstrating the effectiveness of integrating machine learning with cryptographic techniques to secure VANETs. Ultimately, it effectively paves the way for future research into adaptive security mechanisms that can dynamically respond to evolving cyber threats.

An interpretable and transparent machine learning framework for appendicitis detection in pediatric patients.

Appendicitis, an infection and inflammation of the appendix is a prevalent condition in children that requires immediate treatment. Rupture of the appendix may lead to several complications, such as peritonitis and sepsis. Appendicitis is medically diagnosed using urine, blood, and imaging tests. In recent times, Artificial Intelligence and machine learning have been a boon for medicine. Hence, several supervised learning techniques have been utilized in this research to diagnose appendicitis in pediatric patients. Six heterogeneous searching techniques have been used to perform hyperparameter tuning and optimize predictions. These are Bayesian Optimization, Hybrid Bat Algorithm, Hybrid Self-adaptive Bat Algorithm, Firefly Algorithm, Grid Search, and Randomized Search. Further, nine classification metrics were utilized in this study. The Hybrid Bat Algorithm technique performed the best among the above algorithms, with an accuracy of 94% for the customized APPSTACK model. Five explainable artificial intelligence techniques have been tested to interpret the results made by the classifiers. According to the explainers, length of stay, means vermiform appendix detected on ultrasonography, white blood cells, and appendix diameter were the most crucial markers in detecting appendicitis. The proposed system can be used in hospitals for an early/quick diagnosis and to validate the results obtained by other diagnostic modalities.

Chinese word segmentation with heterogeneous graph convolutional network

Abstract Recently, deep learning methods have achieved remarkable success in the Chinese word segmentation (CWS) task. Some of them enhance the CWS model by utilizing contextual features and external resources (e.g., sub-words, lexicon, and syntax). However, existing approaches fail to fully use the heterogeneous features and their structural information. Therefore, in this paper, we propose a heterogeneous information learning framework for CWS, named heterogeneous graph neural segmenter (HGNSeg), which exploits heterogeneous features with the graph convolutional networks and the pretrained language model. Experimental results on six benchmark datasets (e.g., SIGHAN 2005 and SIGHAN 2008) confirm that HGNSeg can effectively improve the performance of CWS. Importantly, HGNSeg also demonstrates an excellent ability to alleviate the out-of-vocabulary (OOV) issue in cross-domain scenarios.

Energy-Saving Multi-Agent Deep Reinforcement Learning Algorithm for Drone Routing Problem.

With the rapid advancement of drone technology, the efficient distribution of drones has garnered significant attention. Central to this discourse is the energy consumption of drones, a critical metric for assessing energy-efficient distribution strategies. Accordingly, this study delves into the energy consumption factors affecting drone distribution. A primary challenge in drone distribution lies in devising optimal, energy-efficient routes for drones. However, traditional routing algorithms, predominantly heuristic-based, exhibit certain limitations. These algorithms often rely on heuristic rules and expert knowledge, which can constrain their ability to escape local optima. Motivated by these shortcomings, we propose a novel multi-agent deep reinforcement learning algorithm that integrates a drone energy consumption model, namely EMADRL. The EMADRL algorithm first formulates the drone routing problem within a multi-agent reinforcement learning framework. It subsequently designs a strategy network model comprising multiple agent networks, tailored to address the node adjacency and masking complexities typical of multi-depot vehicle routing problem. Training utilizes strategy gradient algorithms and attention mechanisms. Furthermore, local and sampling search strategies are introduced to enhance solution quality. Extensive experimentation demonstrates that EMADRL consistently achieves high-quality solutions swiftly. A comparative analysis against contemporary algorithms reveals EMADRL's superior energy efficiency, with average energy savings of 5.96% and maximum savings reaching 12.45%. Thus, this approach offers a promising new avenue for optimizing energy consumption in last-mile distribution scenarios.

Attention Induced Dual Convolutional-Capsule Network (AIDC-CN): A deep learning framework for motor imagery classification

In recent times, Electroencephalography (EEG)-based motor imagery (MI) decoding has garnered significant attention due to its extensive applicability in healthcare, including areas such as assistive robotics and rehabilitation engineering. Nevertheless, the decoding of EEG signals presents considerable challenges owing to their inherent complexity, non-stationary characteristics, and low signal-to-noise ratio. Notably, deep learning-based classifiers have emerged as a prominent focus for addressing the EEG signal decoding process. This study introduces a novel deep learning classifier named the Attention Induced Dual Convolutional-Capsule Network (AIDC-CN) with the specific aim of accurately categorizing various motor imagination class labels. To enhance the classifier’s performance, a dual feature extraction approach leveraging spectrogram and brain connectivity networks has been employed, diversifying the feature set in the classification task. The main highlights of the proposed AIDC-CN classifier includes the introduction of a dual convolution layer to handle the brain connectivity and spectrogram features, addition of a novel self-attention module (SAM) to accentuate the relevant parts of the convolved spectrogram features, introduction of a new cross-attention module (CAM) to refine the outputs obtained from the dual convolution layers and incorporation of a Gaussian Error Linear Unit (GELU) based dynamic routing algorithm to strengthen the coupling among the primary and secondary capsule layers. Performance analysis undertaken on four public data sets depict the superior performance of the proposed model with respect to the state-of-the-art techniques. The code for this model is available at https://github.com/RiteshSurChowdhury/AIDC-CN.

Autonomous driving decision-making control algorithm based on hierarchical reinforcement learning

In the decision-making model of autonomous driving based on learning, the reinforcement learning model has the problems of slow convergence and single application scenario, while the imitation learning model has the problem of poor generalization. In order to solve this problem, a two-layer reinforcement learning framework is proposed to replace the decision layer and control layer in the autonomous driving task. The decision layer divides the driving behavior into lane keeping, left lane change and right lane change. After the decision layer selects the corresponding behavior, it completes the behavior by changing the input of the control layer. Then, a new method for training the control layer, RL_COE (Reinforcement learning combined with online experts), is proposed by combining reinforcement learning and online experts. Finally, a highway simulation environment is built in Carla to verify the proposed algorithm and compare it with the reinforcement learning baseline algorithm. The results show that this method greatly improves the convergence speed and stability of the algorithm and can better complete the driving task.

A platform-agnostic deep reinforcement learning framework for effective Sim2Real transfer towards autonomous driving

Autonomous driving presents unique challenges, particularly in transferring agents trained in simulation to real-world environments due to the discrepancies between the two. To address this issue, here we propose a robust Deep Reinforcement Learning (DRL) framework that incorporates platform-dependent perception modules to extract task-relevant information, enabling the training of a lane-following and overtaking agent in simulation. This framework facilitates the efficient transfer of the DRL agent to new simulated environments and the real world with minimal adjustments. We assess the performance of the agent across various driving scenarios in both simulation and the real world, comparing it to human drivers and a proportional-integral-derivative (PID) baseline in simulation. Additionally, we contrast it with other DRL baselines to clarify the rationale behind choosing this framework. Our proposed approach helps bridge the gaps between different platforms and the Simulation to Reality (Sim2Real) gap, allowing the trained agent to perform consistently in both simulation and real-world scenarios, effectively driving the vehicle.

Representation of Anticipated Rewards and Punishments in the Human Brain.

Subjective value is a core concept in neuroeconomics, serving as the basis for decision making. Despite the extensive literature on the neural encoding of subjective reward value in humans, the neural representation of punishment value remains relatively understudied. This review synthesizes current knowledge on the neural representation of reward value, including methodologies, involved brain regions, and the concept of common currency representation of diverse reward types in decision-making and learning processes. We then critically examine existing research on the neural representation of punishment value, highlighting conceptual and methodological challenges in human studies and insights gained from animal research. Finally, we explore how individual differences in reward and punishment processing may be linked to various mental illnesses, with a focus on stress-related psychopathologies. This review advocates for the integration of both rewards and punishments within value-based decision-making and learning frameworks, leveraging insights from cross-species studies and utilizing ecological gamified paradigms to reflect real-life scenarios.

Machine learning framework for predicting water quality classification

ABSTRACT Groundwater serves as the source for nearly half of the world's drinking water, yet understanding of global groundwater resources remains incomplete, and management of aquifers falls short, particularly concerning groundwater quality. This research offers insights into the groundwater quality in 242 stations of Maharashtra and Union Territory of Dadra and Nagar Haveli and nine parameters (pH, TDS, TH, Calcium (Ca2+), Magnesium (Mg2+), Chloride (Cl−), Sulphate (SO42−), Nitrate (NO3−), Fluoride (F−)) were considered for computing the Water Quality Index (WQI) and hence Water Quality Classification (WQC) based on Water Quality Index (WQI). This research introduces the utilisation of Machine Learning (ML) models, specifically, Random Forest, Adaptive Boosting (AdaBoost), Gradient Boosting, XGBoost, Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) model for predicting WQC and models are tested. Grid search method as a hyperparameter tuning of parameters is utilized to achieve the best possible performance of ML models. The performance metrics that are used for evaluating and reporting the performance of classification models are Accuracy, Precision, Recall or Sensitivity, F1 Score. SVM achieved the highest performance in predicting WQC. With accurate predictions of WQC, these findings have the potential to enhance NEP concerning water resources by facilitating ongoing improvements in water quality.

A multi-modality and multi-granularity collaborative learning framework for identifying spatial domains and spatially variable genes.

Recent advances in spatial transcriptomics technologies have provided multi-modality data integrating gene expression, spatial context, and histological images. Accurately identifying spatial domains and spatially variable genes is crucial for understanding tissue structures and biological functions. However, effectively combining multi-modality data to identify spatial domains and determining SVGs closely related to these spatial domains remains a challenge. In this study, we propose spatial transcriptomics Multi-modality and Multi-granularity Collaborative Learning (spaMMCL). For detecting spatial domains, spaMMCL mitigates the adverse effects of modality bias by masking portions of gene expression data, integrates gene and image features using a shared graph convolutional network, and employs graph self-supervised learning to deal with noise from feature fusion. Simultaneously, based on the identified spatial domains, spaMMCL integrates various strategies to detect potential SVGs at different granularities, enhancing their reliability and biological significance. Experimental results demonstrate that spaMMCL substantially improves the identification of spatial domains and SVGs. The code and data of spaMMCL are available on Github: Https://github.com/liangxiao-cs/spaMMCL. Supplementary data are available at Bioinformatics online.

Evaluating Airport Service Quality Based on the Statistical and Predictive Analysis of Skytrax Passenger Reviews

This study leverages approximately 7500 reviews from Skytrax to explore the determinants of airport service quality and their influence on passenger recommendations. The dataset includes various features such as terminal cleanliness, terminal seating, terminal signs, food and beverages, airport shopping, WiFi connectivity, and airport staff. The research employs a comprehensive methodology encompassing statistical data analysis, predictive modelling, and interaction effects analysis. The descriptive analysis of time-series data highlighted trends and fluctuations in service quality and recommendations, providing insights into temporal dynamics. Multiple machine learning models, including logistic regression, Random Forest, SVM, KNN, Gradient Boosting, and Neural Networks, were developed in this study and cross-validated for airport recommendation based on Skytrax’s online reviews. Among others, Gradient Boosting emerged as the most accurate model with an 88.15% mean accuracy. Interaction effects revealed significant combined influences, such as terminal cleanliness and terminal seating, on passenger recommendations. This multifaceted approach offers robust insights into factors influencing airport recommendations and guides improvements in airport management to enhance passenger satisfaction. Future work will focus on a general-purpose machine learning framework and its toolbox development for airport service quality analysis based on online reviews from various sources.

SMAR: self-supervised mobile application recommendation based on graph convolutional networks

PurposeWith the exponential growth of mobile applications, recommending suitable mobile applications to users becomes a critical challenge. Although existing methods have made achievements in mobile application recommendation by leveraging graph convolutional networks (GCNs), they suffer from two limitations: the reliance on a singular acquisition path leads to signal sparsity, and the neighborhood aggregation method exacerbates the adverse impact of noisy interactions. This paper aims to propose SMAR, a self-supervised mobile application recommendation approach based on GCN, which is designed to overcome existing challenges by using self-supervised learning to create an auxiliary task.Design/methodology/approachIn detail, this method uses three distinct data augmentation techniques node dropout, edge dropout and random walk, which create varied perspectives of each node. Then compares these perspectives, aiming to ensure uniformity across different views of the same node while maintaining the differences between separate nodes. Ultimately, auxiliary task is combined with the primary supervised task using a multi-task learning framework, thereby refining the overall mobile application recommendation process.FindingsExtensive experiments on two real datasets demonstrate that SMAR achieves better Recall and NDCG performances than other strong baselines, validating the effectiveness of the proposed method.Originality/valueIn this paper, the authors introduce self-supervised learning into mobile application recommendation approach based on GCNs. This method enhances traditional supervised tasks by using auxiliary task to provide additional information, thereby improving signal accuracy and reducing the influence of noisy interactions in mobile application recommendations.

Low-complexity end-to-end deep learning framework for 100G-PON

End-to-end learning allows communication systems to achieve optimal performance compared with conventional blockwise structure design. By modeling the channel with neural networks and training the transmitter and receiver on this differentiable channel, the whole system can be jointly optimized. However, in existing schemes, channel modeling methods, such as the generative adversarial network and long short-term memory network, have complex architectures and cannot track channel changes, leading to less effective end-to-end learning. Meanwhile, the complexity of neural networks deployed at the transmitter and receiver is too high for practical applications. In this work, we propose an efficient and low-complexity end-to-end deep learning framework and experimentally validate it on a 100G passive optical network. It uses a noise adaptation network to model channel response and noise distribution and employs offline pretraining and online tracking training to improve the efficiency and accuracy of channel modeling. For the transmitter, it consists of a pattern-dependent look-up table (PDLUT) based on a neural network (NN-PDLUT) with a single convolutional layer. Further, the receiver is also an NN with a single convolutional layer; thus, the end-to-end signal processing is extremely simple. The experimental results show that end-to-end learning improves the receiver sensitivity by 0.85 and 1.59 dB compared with receiver-only equalization based on Volterra nonlinear equalization (VNLE) and joint equalization based on a PDLUT and a feed-forward equalizer, respectively. Moreover, the number of multiply–accumulate operations consumed by the transmitter and receiver in the end-to-end learning scheme is reduced by 75.7% compared with VNLE-based receiver-only equalization.

Learning Robust Sequential Recommenders through Confident Soft Labels

Sequential recommenders that are trained on implicit feedback are usually learned as a multi-class classification task through softmax-based loss functions on one-hot class labels. However, one-hot training labels are sparse and may lead to biased training and sub-optimal performance. Dense, soft labels have been shown to help improve recommendation performance. However, how to generate high-quality and confident soft labels from noisy sequential interactions between users and items is still an open question. We propose a new learning framework for sequential recommenders, CSRec, which introduces c onfident s oft labels to provide robust guidance when learning from user-item interactions. CSRec contains a teacher module that generates high-quality and confident soft labels and a student module that acts as the target recommender and is trained on the combination of dense, soft labels and sparse, one-hot labels. We propose and compare three approaches to constructing the teacher module: (i) model-level, (ii) data-level, and (iii) training-level. To evaluate the effectiveness and generalization ability of CSRec, we conduct experiments using various state-of-the-art sequential recommendation models as the target student module on four benchmark datasets. Our experimental results demonstrate that CSRec is effective in training better-performing sequential recommenders.

Predictive Modeling of Preeclampsia Risk Using Random Forest Algorithm within a Machine Learning Framework

Preeclampsia is a serious pregnancy complication characterized by high blood pressure, potentially leading to organ damage, making early risk prediction crucial to reducing maternal morbidity and mortality. This study aims to develop a preeclampsia risk prediction model using medical and clinical data from 80 patients at Rumah Bersalin Sadan. The data include demographic profiles, blood pressure, weight, maternal age, preeclampsia history, body mass index, number of previous pregnancies, as well as genetic and environmental factors. The dependent variable is the risk of preeclampsia, either as a binary outcome (yes/no) or as a continuous risk score. The predictive model was built using multivariate linear regression and the Random Forest algorithm. The results showed that the Random Forest model achieved an accuracy of 65.22%, with an F-statistic of 7.345 and a very small p-value (1.908e-06), indicating that the model effectively explains data variability. However, the low Kappa value suggests room for improvement through feature refinement, hyperparameter tuning, or exploring other algorithms. Although these findings suggest that Random Forest is a promising method, further evaluation and model optimization are needed to enhance predictive performance and determine whether this method is the most suitable for the dataset used.

Knowledge‐Based Deep Learning to Predict Vegetation Carbon, Nitrogen and Phosphorus Densities in China’s Shrublands

AbstractAccurate estimations of carbon (C), nitrogen (N), and phosphorus (P) densities in shrublands are pivotal for assessing terrestrial ecosystem carbon sequestration. Combining in‐situ investigations and machine learning facilitates large‐scale patterns mapping, however, which often overlooks underlying ecological regulations. Here we utilize data from 1,122 survey plots across China's shrublands and develop a novel knowledge‐based deep learning framework that integrates a structural equation model (SEM) to elucidate mechanisms and construct an artificial neural network (ANN) based on these causal relationships. Results show that biomass allocation to different organs follows allometric regulations and that N and P concentrations maintain a degree of stoichiometric homeostasis following biological stoichiometry theory. This insight guides the construction of our ANN, which outperforms both SEM and other prevalent machine learning methods. By leveraging ecological theories to inform model construction, our framework not only enhances prediction accuracy and explainability but also provides a methodological blueprint for ecological research.

Advancing remote sensing: a unified deep learning approach with pretrained and custom architectures for high-precision classification

Advancement of remote sensing is vital for accurate land cover mapping and ecological surveillance. This research proposes a cutting-edge deep learning framework that integrates a meticulously designed customary network architecture with the effectiveness of pre-trained models, namely GoogleNet, VGG16, and InceptionV3. Our approach captures the complex features of RGB satellite images throughout various land cover categories using diverse EuroSAT datasets. Our system achieves an impressive 99.40% test accuracy by implementing feature extraction through effective models with a customized deep learning network in an intuitive manner. High class-wise accuracies are observed, ranging from 96.00% to 100.00%, whereas F1-scores, precision, and recall all converge at 99.40%. The aforementioned results demonstrate our approach’s perspective to significantly improve analysis while also confirming its superiority. These results not only confirm the performance of our algorithm but also show that it could be used to redesign satellite visual analysis behavior. These results open the window to further improve and accurate remote sensing systems, giving a helpful viewpoint for researchers as well as professionals.