Learning Models For Tasks Research Articles

Regional choroidal thickness estimation from color fundus images based on convolutional neural networks

PurposeThis study aims to estimate the regional choroidal thickness from color fundus images from convolutional neural networks in different network structures and task learning models. Method1276 color fundus photos and their corresponding choroidal thickness values from healthy subjects were obtained from the Topcon DRI Triton optical coherence tomography machine. Initially, ten commonly used convolutional neural networks were deployed to identify the most accurate model, which was subsequently selected for further training. This selected model was then employed in combination with single-, multiple-, and auxiliary-task training models to predict the average and sub-region choroidal thickness in both ETDRS (Early Treatment Diabetic Retinopathy Study) grids and 100-grid subregions. The values of mean absolute error and coefficient of determination (R2) were involved to evaluate the models' performance. ResultsEfficientnet-b0 network outperformed other networks with the lowest mean absolute error value (25.61 μm) and highest R2 (0.7817) in average choroidal thickness. Incorporating diopter spherical, anterior chamber depth, and lens thickness as auxiliary tasks improved predicted accuracy (p-value = 6.39×10−44, 2.72×10−38, 1.15×10−36 respectively). For ETDRS regional choroidal thickness estimation, multi-task model achieved better results than single task model (lowest mean absolute error = 31.10 μm vs. 33.20 μm). The multi-task training also can simultaneously predict the choroidal thickness of 100 grids with a minimum mean absolute error of 33.86 μm. ConclusionsEfficientnet-b0, in combination with multi-task and auxiliary task models, achieve high accuracy in estimating average and regional macular choroidal thickness directly from color fundus photographs.

Survey of transformers and towards ensemble learning using transformers for natural language processing

The transformer model is a famous natural language processing model proposed by Google in 2017. Now, with the extensive development of deep learning, many natural language processing tasks can be solved by deep learning methods. After the BERT model was proposed, many pre-trained models such as the XLNet model, the RoBERTa model, and the ALBERT model were also proposed in the research community. These models perform very well in various natural language processing tasks. In this paper, we describe and compare these well-known models. In addition, we also apply several types of existing and well-known models which are the BERT model, the XLNet model, the RoBERTa model, the GPT2 model, and the ALBERT model to different existing and well-known natural language processing tasks, and analyze each model based on their performance. There are a few papers that comprehensively compare various transformer models. In our paper, we use six types of well-known tasks, such as sentiment analysis, question answering, text generation, text summarization, name entity recognition, and topic modeling tasks to compare the performance of various transformer models. In addition, using the existing models, we also propose ensemble learning models for the different natural language processing tasks. The results show that our ensemble learning models perform better than a single classifier on specific tasks.Graphical

Evaluating the Efficacy of Traditional Machine Learning Models in Speaker Recognition: A Comparative Study Using the LibriSpeech Dataset

The efficacy of machine learning models in speaker recognition tasks is critical for advancements in security systems, biometric authentication, and personalized user interfaces. This study provides a comparative analysis of three prominent machine learning models: Naive Bayes, Logistic Regression, and Gradient Boosting, using the LibriSpeech test-clean dataset—a corpus of read English speech from audiobooks designed for training and evaluating speech recognition systems. Mel-Frequency Cepstral Coefficients (MFCCs) were extracted as features from the audio samples to represent the power spectrum of the speakers’ voices. The models were evaluated based on precision, recall, F1-score, and accuracy to determine their performance in correctly identifying speakers. Results indicate that Logistic Regression outperformed the other models, achieving nearly perfect scores across all metrics, suggesting its superior capability for linear classification in high-dimensional spaces. Naive Bayes also demonstrated high efficiency and robustness, despite the inherent assumption of feature independence, while Gradient Boosting showed slightly lower performance, potentially due to model complexity and overfitting. The study underscores the potential of simpler machine learning models to achieve high accuracy in speaker recognition tasks, particularly where computational resources are limited. However, limitations such as the controlled nature of the dataset and the focus on a single feature type were noted, with recommendations for future research to include more diverse environmental conditions and feature sets.

MHAiR: A Dataset of Audio-Image Representations for Multimodal Human Actions

Audio-image representations for a multimodal human action (MHAiR) dataset contains six different image representations of the audio signals that capture the temporal dynamics of the actions in a very compact and informative way. The dataset was extracted from the audio recordings which were captured from an existing video dataset, i.e., UCF101. Each data sample captured a duration of approximately 10 s long, and the overall dataset was split into 4893 training samples and 1944 testing samples. The resulting feature sequences were then converted into images, which can be used for human action recognition and other related tasks. These images can be used as a benchmark dataset for evaluating the performance of machine learning models for human action recognition and related tasks. These audio-image representations could be suitable for a wide range of applications, such as surveillance, healthcare monitoring, and robotics. The dataset can also be used for transfer learning, where pre-trained models can be fine-tuned on a specific task using specific audio images. Thus, this dataset can facilitate the development of new techniques and approaches for improving the accuracy of human action-related tasks and also serve as a standard benchmark for testing the performance of different machine learning models and algorithms.

Continually adapting pre-trained language model to universal annotation of single-cell RNA-seq data.

Cell-type annotation of single-cell RNA-sequencing (scRNA-seq) data is a hallmark of biomedical research and clinical application. Current annotation tools usually assume the simultaneous acquisition of well-annotated data, but without the ability to expand knowledge from new data. Yet, such tools are inconsistent with the continuous emergence of scRNA-seq data, calling for a continuous cell-type annotation model. In addition, by their powerful ability of information integration and model interpretability, transformer-based pre-trained language models have led to breakthroughs in single-cell biology research. Therefore, the systematic combining of continual learning and pre-trained language models for cell-type annotation tasks is inevitable. We herein propose a universal cell-type annotation tool, called CANAL, that continuously fine-tunes a pre-trained language model trained on a large amount of unlabeled scRNA-seq data, as new well-labeled data emerges. CANAL essentially alleviates the dilemma of catastrophic forgetting, both in terms of model inputs and outputs. For model inputs, we introduce an experience replay schema that repeatedly reviews previous vital examples in current training stages. This is achieved through a dynamic example bank with a fixed buffer size. The example bank is class-balanced and proficient in retaining cell-type-specific information, particularly facilitating the consolidation of patterns associated with rare cell types. For model outputs, we utilize representation knowledge distillation to regularize the divergence between previous and current models, resulting in the preservation of knowledge learned from past training stages. Moreover, our universal annotation framework considers the inclusion of new cell types throughout the fine-tuning and testing stages. We can continuously expand the cell-type annotation library by absorbing new cell types from newly arrived, well-annotated training datasets, as well as automatically identify novel cells in unlabeled datasets. Comprehensive experiments with data streams under various biological scenarios demonstrate the versatility and high model interpretability of CANAL. An implementation of CANAL is available from https://github.com/aster-ww/CANAL-torch. dengmh@pku.edu.cn. Supplementary data are available at Journal Name online.

Effective Approach for Fine-Tuning Pre-Trained Models for the Extraction of Texts From Source Codes

This study introduces SR-Text, a robust approach leveraging pre-trained models like BERT and T5 for enhanced text extraction from source codes. Addressing the limitations of traditional manual summarization, our methodology focuses on fine-tuning these models to better understand and generate contextual summaries, thus overcoming challenges such as long-term dependency and dataset quality issues. We conduct a detailed analysis of programming language syntax and semantics to develop syntax-aware text retrieval techniques, significantly boosting the accuracy and relevance of the texts extracted. The paper also explores a hybrid approach that integrates statistical machine learning with rule-based methods, enhancing the robustness and adaptability of our text extraction processes across diverse coding styles and languages. Empirical results from a meticulously curated dataset demonstrate marked improvements in performance metrics: precision increased by 15%, recall by 20%, and an F1 score enhancement of 18%. These improvements underscore the effectiveness of using advanced machine learning models in software engineering tasks. This research not only paves the way for future work in multilingual code summarization but also discusses broader implications for automated software analysis tools, proposing directions for future research to further refine and expand this methodology.

Enhancing Robustness and Generalization in Deep Learning Models for Image Processing

In recent years, deep learning models have demonstrated remarkable success in various image processing tasks, ranging from object recognition to medical image analysis. However, their performance often degrades in the presence of unseen data or adversarial attacks, highlighting the need for enhancing robustness and generalization. This paper explores innovative approaches to address these challenges, aiming to improve the reliability and applicability of deep learning models in real-world scenarios. The first section of the paper delves into the importance of robustness and generalization in deep learning models for image processing tasks. It discusses the implications of model vulnerabilities, such as overfitting to training data and susceptibility to adversarial perturbations, on the reliability of model predictions. [1] Through a comprehensive review of existing literature, various factors influencing robustness and generalization are identified, including dataset diversity, model architecture, regularization techniques, and adversarial training methods. The paper proposes novel methodologies to enhance the robustness and generalization capabilities of deep learning models. One key approach involves the integration of diverse training data sources, including synthetic data augmentation and domain adaptation techniques, to expose the model to a wider range of scenarios and improve its ability to generalize to unseen data. Additionally, advanced regularization techniques, such as dropout and batch normalization, are explored to mitigate overfitting and improve model generalization. The paper investigates the effectiveness of adversarial training strategies in enhancing model robustness against adversarial attacks. By incorporating adversarially generated examples during training, deep learning models can learn to better resist perturbations and maintain performance under adversarial conditions. Moreover, the paper explores the potential of incorporating uncertainty estimation methods, such as Bayesian neural networks and Monte Carlo dropout, to quantify model uncertainty and improve robustness in uncertain environments. This paper presents a comprehensive investigation into enhancing robustness and generalization in deep learning models for image processing tasks. By addressing key challenges such as overfitting, dataset bias, and adversarial vulnerabilities, the proposed methodologies offer promising avenues for improving the reliability and applicability of deep learning models in real-world scenarios.

Interpretable extractive text summarization with meta-learning and BI-LSTM: A study of meta learning and explainability techniques

Text summarization is a widely-researched problem among scholars in the field of natural language processing. Multiple techniques have been proposed to help tackle this problem, yet some of these methodologies may still exhibit limitations such as the requirements for large training datasets, which might not always be possible, but more importantly, the lack of interpretability or transparency of the model. In this paper, we propose using meta-learning algorithm to train a deep learning model for extractive text summarization and then using various explanatory techniques such as SHAP (Shapley, 1953), linear regression (Lederer, 2022), decision trees (Fürnkranz, 2010), and input modification to gain insights into the model’s decision making process. To evaluate the effectiveness of our approach, we will compare it to other popular natural language processing models like BERT (Miller, 2019) or XLNET (Yang et al., 2020) using the ROUGE metrics (Lin, 2004). Overall, our proposed approach provides a promising solution to the limitations of existing methods and a framework for improving the explainability of deep learning models in other natural language processing tasks.

Explainable End-to-End Supervised Learning Identifies Myelodysplastic Neoplasms in Bone Marrow Smears

Myelodysplastic neoplasms (MDS) encompass clonal myeloid malignancies that are characterized by ineffective hematopoiesis, cytopenia, myelodysplasia, and recurrent genetic events. Apart from genetic abnormalities and the assessment of peripheral blood for cytopenia, cytomorphologic evaluation of bone marrow morphology is key in initial diagnosis and response assessment as well as to detect disease transformation to acute myeloid leukemia (AML). However, the identification of aberrant morphologies in bone marrow cells is challenging and prone to inter-observer variability even for seasoned morphologists resulting in discrepancies between local site and central review. After the introduction of deep learning (DL) models for computer vision tasks, a wide variety of applications in imaging diagnostics has been addressed using convolutional neural networks including the assessment of bone marrow cell types and prediction of hematological malignancies. Bone marrow smear (BMS) images of manually selected regions of interest (ROI) from 483 MDS patients treated at the University Hospital Dresden were captured at 40-fold magnification with a resolution of 2560*1920 pixels and pre-processed. In parallel, BMS images from two control cohorts were taken: (i) 1226 AML patients registered in the German Study Alliance Leukemia (SAL) registry, and (ii) 236 healthy bone marrow donors. In addition, an external validation cohort of 50 MDS patients was obtained from the Munich Leukemia Laboratory (MLL). Image-level labels were provided denoting the corresponding diagnosis: “MDS”, “AML” and “healthy donor”. To balance different sample sizes between the datasets, simple image augmentation techniques such as random sized cropping, color shifting, and linear transformations were used. Training and testing were performed with 5-fold cross-validation, i.e. multiple train-test-splits of 80:20. Importantly, cell-level labels, e.g. specifically annotated cell types or features like distinct dysplastic morphologies were not used to train our model as we intended to abstract visual information exclusively through our neural networks instead of using expert-guided feature engineering. Six different recently introduced convolutional neural network (CNN) architectures were compared (ResNet-34/50/101/152, ResNeXt-50/101, ShuffleNet V2, DenseNet-121/169/201, Wide ResNet-50/101, and SqueezeNet 1.1). To evaluate individual model performance, we used accuracy, recall (=sensitivity), precision (=positive predictive value), and area-under-the-curve (AUC) of both the receiver-operating-characteristic (ROC) and precision-recall-curve (PRC). Binary classification tasks showed high accuracy of 0.956 and 0.981 for distinguishing MDS from healthy donors and AML, respectively (Table I). Results were validated in the external MDS cohort reaching an accuracy of 0.916 for MDS vs. donors and 0.968 for MDS vs. AML (Table I). For explainability, we used occlusion sensitivity mapping which iteratively blocks image areas from being evaluated by the CNNs, measures consecutive drops in prediction performance, and thus identifies image areas of high importance (highlighted in green). By reviewing images, we found the CNNs to target nuclei not only of dysplastic cells but rather of all available cells in a ROI (Figure 1). Using supervised end-to-end deep learning, we developed a software framework to distinguish between MDS, AML, and healthy controls with very high accuracy. Importantly, models do not require laborious feature engineering, i.e. manually drafted, cell-level labels alleviating the need for time-consuming and cost-ineffective manual labeling and thereby bypassing a key bottleneck for current standards of computer vision in microscopy.

Fractal Dimension as a discriminative feature for high accuracy classification in motor imagery EEG-based brain-computer interface

Background and ObjectiveThe brain-computer interface (BCI) technology acquires human brain electrical signals, which can be effectively and successfully used to control external devices, potentially supporting subjects suffering from motor impairments in the interaction with the environment. To this aim, BCI systems must correctly decode and interpret neurophysiological signals reflecting the intention of the subjects to move. Therefore, the accurate classification of single events in motor tasks represents a fundamental challenge in ensuring efficient communication and control between users and BCIs. Movement-associated changes in electroencephalographic (EEG) sensorimotor rhythms, such as event-related desynchronization (ERD), are well-known features of discriminating motor tasks. Fractal dimension (FD) can be used to evaluate the complexity and self-similarity in brain signals, potentially providing complementary information to frequency-based signal features. MethodsIn the present work, we introduce FD as a novel feature for subject-independent event classification, and we test several machine learning (ML) models in behavioural tasks of motor imagery (MI) and motor execution (ME). ResultsOur results show that FD improves the classification accuracy of ML compared to ERD. Furthermore, unilateral hand movements have higher classification accuracy than bilateral movements in both MI and ME tasks. ConclusionsThese results provide further insights into subject-independent event classification in BCI systems and demonstrate the potential of FD as a discriminative feature for EEG signals.

Reducing annotation burden in MR: A novel MR-contrast guided contrastive learning approach for image segmentation.

Contrastive learning, a successful form of representational learning, has shown promising results in pretraining deep learning (DL) models for downstream tasks. When working with limited annotation data, as in medical image segmentation tasks, learning domain-specific local representations can further improve the performance of DL models. In this work, we extend the contrastive learning framework to utilize domain-specific contrast information from unlabeled Magnetic Resonance (MR) images to improve the performance of downstream MR image segmentation tasks in the presence of limited labeled data. The contrast in MR images is controlled by underlying tissue properties (e.g., T1 or T2) and image acquisition parameters. We hypothesize that learning to discriminate local representations based on underlying tissue properties should improve subsequent segmentation tasks on MR images. We propose a novel constrained contrastive learning (CCL) strategy that uses tissue-specific information via a constraint map to define positive and negative local neighborhoods for contrastive learning, embedding this information in the representational space during pretraining. For a given MR contrast image, the proposed strategy uses local signal characteristics (constraint map) across a set of related multi-contrast MR images as a surrogate for underlying tissue information. We demonstrate the utility of the approach for downstream: (1) multi-organ segmentation tasks in T2-weighted images where a DL model learns T2 information with constraint maps from a set of 2D multi-echo T2-weighted images (n=101) and (2) tumor segmentation tasks in multi-parametric images from the public brain tumor segmentation (BraTS) (n=80) dataset where DL models learn T1 and T2 information from multi-parametric BraTS images. Performance is evaluated on downstream multi-label segmentation tasks with limited data in (1) T2-weighted images of the abdomen from an in-house Radial-T2 (Train/Test=30/20), (2) public Cartesian-T2 (Train/Test=6/12) dataset, and (3) multi-parametric MR images from the public brain tumor segmentation dataset (BraTS) (Train/Test=40/50). The performance of the proposed CCL strategy is compared to state-of-the-art self-supervised contrastive learning techniques. In each task, a model is also trained using all available labeled data for supervised baseline performance. The proposed CCL strategy consistently yielded improved Dice scores, Precision, and Recall metrics, and reduced HD95 values across all segmentation tasks. We also observed performance comparable to the baseline with reduced annotation effort. The t-SNE visualization of features for T2-weighted images demonstrates its ability to embed T2 information in the representational space. On the BraTS dataset, we also observed that using an appropriate multi-contrast space to learn T1+T2, T1, or T2 information during pretraining further improved the performance of tumor segmentation tasks. Learning to embed tissue-specific information that controls MR image contrast with the proposed constrained contrastive learning improved the performance of DL models on subsequent segmentation tasks compared to conventional self-supervised contrastive learning techniques. The use of such domain-specific local representations could help understand, improve performance, and mitigate the scarcity of labeled data in MR image segmentation tasks.

A Graph-Based Context-Aware Model to Understand Online Conversations

Online forums that allow for participatory engagement between users have been transformative for the public discussion of many important issues. However, such conversations can sometimes escalate into full-blown exchanges of hate and misinformation. Existing approaches in natural language processing (NLP), such as deep learning models for classification tasks, use as inputs only a single comment or a pair of comments depending upon whether the task concerns the inference of properties of the individual comments or the replies between pairs of comments, respectively. However, in online conversations, comments and replies may be based on external context beyond the immediately relevant information that is input to the model. Therefore, being aware of the conversations’ surrounding contexts should improve the model’s performance for the inference task at hand. We propose GraphNLI , 1 a novel graph-based deep learning architecture that uses graph walks to incorporate the wider context of a conversation in a principled manner. Specifically, a graph walk starts from a given comment and samples “nearby” comments in the same or parallel conversation threads, which results in additional embeddings that are aggregated together with the initial comment’s embedding. We then use these enriched embeddings for downstream NLP prediction tasks that are important for online conversations. We evaluate GraphNLI on two such tasks - polarity prediction and misogynistic hate speech detection - and find that our model consistently outperforms all relevant baselines for both tasks. Specifically, GraphNLI with a biased root-seeking random walk performs with a macro- F 1 score of 3 and 6 percentage points better than the best-performing BERT-based baselines for the polarity prediction and hate speech detection tasks, respectively. We also perform extensive ablative experiments and hyperparameter searches to understand the efficacy of GraphNLI. This demonstrates the potential of context-aware models to capture the global context along with the local context of online conversations for these two tasks.

Effect of the deep residual networks at different depths on expression recognition

Facial expression recognition plays a critical role in numerous applications like emotion analysis, human-computer interaction, and surveillance systems. Given the importance of this task, this study aims to investigate the effectiveness of different depths of Residual Networks (ResNet). The primary objective is to scrutinize and compare these ResNet models in terms of their training and validation losses and performance metrics like accuracy, recall, and F1 scores. In this research, a thorough comparative analysis is conducted by setting up exhaustive experiments using these models. The experiment is carried out on a popular facial expression dataset. Despite the depth differences, ResNet101 emerged as the model demonstrating superior performance. It struck the most effective balance between model complexity and generalization capacity, leading to the lowest validation loss and better performance. Experimental results show that a more complex model does not necessarily yield optimal results. The optimal balance between model complexity and generalisation needs to be investigated. These findings can provide essential guidance in the design of deep learning models for facial expression recognition and other similar tasks.

Deep Learning Methods for Semantic Segmentation in Remote Sensing with Small Data: A Survey

The annotations used during the training process are crucial for the inference results of remote sensing images (RSIs) based on a deep learning framework. Unlabeled RSIs can be obtained relatively easily. However, pixel-level annotation is a process that necessitates a high level of expertise and experience. Consequently, the use of small sample training methods has attracted widespread attention as they help alleviate reliance on large amounts of high-quality labeled data and current deep learning methods. Moreover, research on small sample learning is still in its infancy owing to the unique challenges faced when completing semantic segmentation tasks with RSI. To better understand and stimulate future research that utilizes semantic segmentation tasks with small data, we summarized the supervised learning methods and challenges they face. We also reviewed the supervised approaches with data that are currently popular to help elucidate how to efficiently utilize a limited number of samples to address issues with semantic segmentation in RSI. The main methods discussed are self-supervised learning, semi-supervised learning, weakly supervised learning and few-shot methods. The solution of cross-domain challenges has also been discussed. Furthermore, multi-modal methods, prior knowledge constrained methods, and future research required to help optimize deep learning models for various downstream tasks in relation to RSI have been identified.

LEAF: Navigating Concept Drift in Cellular Networks

Operational networks commonly rely on machine learning models for many tasks, including detecting anomalies, inferring application performance, and forecasting demand. Yet, model accuracy can degrade due to concept drift, whereby the relationship between the features and the target to be predicted changes. Mitigating concept drift is an essential part of operationalizing machine learning models in general, but is of particular importance in networking's highly dynamic deployment environments. In this paper, we first characterize concept drift in a large cellular network for a major metropolitan area in the United States. We find that concept drift occurs across many important key performance indicators (KPIs), independently of the model, training set size, and time interval---thus necessitating practical approaches to detect, explain, and mitigate it. We then show that frequent model retraining with newly available data is not sufficient to mitigate concept drift, and can even degrade model accuracy further. Finally, we develop a new methodology for concept drift mitigation, Local Error Approximation of Features (LEAF). LEAF works by detecting drift; explaining the features and time intervals that contribute the most to drift; and mitigates it using forgetting and over-sampling. We evaluate LEAF against industry-standard mitigation approaches (notably, periodic retraining) with more than four years of cellular KPI data. Our initial tests with a major cellular provider in the US show that LEAF consistently outperforms periodic and triggered retraining on complex, real-world data while reducing costly retraining operations.

Evading text based emotion detection mechanism via adversarial attacks

Textual Emotion Analysis (TEA) seeks to extract and assess the emotional states of users from the text. Various Deep Learning (DL) algorithms have emerged rapidly and demonstrated success in numerous disciplines, including audio, image, and natural language processing. Thetrend has shifted a growing number of researchers from standard machine learning to DL for scientific study. Using DL approaches, we offer an overview of TEA in this paper. After introducing the background for emotion analysis, including the definition of emotion, emotion classification methods, and application domains of emotion analysis, we demonstrated that, despite the immense success of deep learning models in NLP-related tasks, they are susceptible to adversarial attacks, which can lead to incorrect emotion classification. An adversarial text is constructed by altering a few words or characters so as to keep the overall semantic similarity of emotion for a human reader while tricking the machine into making erroneous predictions. This study demonstrates the vulnerability of emotion categorization by generating adversarial text using a variety of cutting-edge attack techniques. Comprehensive experiments are performed to assess the effectiveness of the attack methods against several widely-used models, such as Word-CNN, Bi-LSTM, and four powerful transformer models, namely BERT, DistilBERT, ALBERT, and RoBERTa. These models were trained on an emotion dataset utilized for the purpose of emotion classification. We evaluated and analyzed the behavior of different models under a variety of attack conditions to determine which is the most and least vulnerable. Also, we determine which perturbation technique affects transformer models the most. Using Attack Success Rates (ASR) as our evaluation metric, we have assessed the potential outcomes. The findings reveal that methodologies for classifying emotion predictioncan be circumvented, which has implications for existing policy measures.

Advances in Data‐Driven Analysis and Synthesis of 3D Indoor Scenes

AbstractThis report surveys advances in deep learning‐based modelling techniques that address four different 3D indoor scene analysis tasks, as well as synthesis of 3D indoor scenes. We describe different kinds of representations for indoor scenes, various indoor scene datasets available for research in the aforementioned areas, and discuss notable works employing machine learning models for such scene modelling tasks based on these representations. Specifically, we focus on the analysis and synthesis of 3D indoor scenes. With respect to analysis, we focus on four basic scene understanding tasks – 3D object detection, 3D scene segmentation, 3D scene reconstruction and 3D scene similarity. And for synthesis, we mainly discuss neural scene synthesis works, though also highlighting model‐driven methods that allow for human‐centric, progressive scene synthesis. We identify the challenges involved in modelling scenes for these tasks and the kind of machinery that needs to be developed to adapt to the data representation, and the task setting in general. For each of these tasks, we provide a comprehensive summary of the state‐of‐the‐art works across different axes such as the choice of data representation, backbone, evaluation metric, input, output and so on, providing an organized review of the literature. Towards the end, we discuss some interesting research directions that have the potential to make a direct impact on the way users interact and engage with these virtual scene models, making them an integral part of the metaverse.

VisGIN: Visibility Graph Neural Network on one-dimensional data for biometric authentication

The research community has extensively utilized biometrics, which is one of the most prominent sources of knowledge for authentication schemes in the rapidly advancing field of cybersecurity. Among various types of biometric information, an individual’s electrocardiogram (ECG) plot serves as an inimitable object as it cannot be faked intentionally. For large-scale applications, where granting access to genuine signal owners is necessary, deep learning is an essential component. In this context, the convergence between one-dimensional representations and Graph Neural Networks (GNN) includes prospective solutions. To capture the time-series data from a different perspective, hereby, we propose the VisGIN model which utilizes GINConv for the convolutional layers and passes Visibility Graphs (VG) as input. In parallel with our intuition, the findings of this study affirmed the fruition of the VisGIN approach, offering significant implications for the field of ECG authentication. By achieving an average classification accuracy of 99.76% in the evaluation of grant-access decisions, our VisGIN model demonstrates the effectiveness of graph machine learning models for time-series binary classification tasks, particularly in ECG authentication. As a result, our study provides a valuable advancement in enhancing the security and reliability of authentication systems. Researchers and practitioners can benefit from our work by leveraging the VisGIN model and its graph-based approach to bolster the accuracy and robustness of ECG authentication systems. Our code is available at https://github.com/AslantheAslan/visibility-gin.

Data-efficient GNN models of communication networks using beta-distribution-based sample ranking

Machine learning models for tasks in communication networks often require large datasets to be trained. This training is cost intensive, and solutions to reduce these costs are required. It is not clear what the best approach to solve this problem is. Here we show an approach that is able to create a minimally-sized training dataset while maintaining high predictive power of the model. We apply our approach to a state-of-the-art graph neural network model for performance prediction in communication networks. Our approach is limited to a dataset of 100 samples with reduced sizes and achieves an MAPE of 9.79% on a test dataset containing significantly larger problem sizes, compared to a baseline approach which achieved an MAPE of 37.82%. We think this approach can be useful to create high-quality datasets of communication networks and decrease the time needed to train graph neural network models on performance prediction tasks.

Automated Optimization-Based Deep Learning Models for Image Classification Tasks

Applying deep learning models requires design and optimization when solving multifaceted artificial intelligence tasks. Optimization relies on human expertise and is achieved only with great exertion. The current literature concentrates on automating design; optimization needs more attention. Similarly, most existing optimization libraries focus on other machine learning tasks rather than image classification. For this reason, an automated optimization scheme of deep learning models for image classification tasks is proposed in this paper. A sequential-model-based optimization algorithm was used to implement the proposed method. Four deep learning models, a transformer-based model, and standard datasets for image classification challenges were employed in the experiments. Through empirical evaluations, this paper demonstrates that the proposed scheme improves the performance of deep learning models. Specifically, for a Virtual Geometry Group (VGG-16), accuracy was heightened from 0.937 to 0.983, signifying a 73% relative error rate drop within an hour of automated optimization. Similarly, training-related parameter values are proposed to improve the performance of deep learning models. The scheme can be extended to automate the optimization of transformer-based models. The insights from this study may assist efforts to provide full access to the building and optimization of DL models, even for amateurs.