Computational Resource Requirements Research Articles

Walking on two legs: Joint service placement and computation configuration for provisioning containerized services at edges

With the development of edge computing and container virtualization, provisioning containerized services at network edges is proposed for high responsiveness and low wide area network (WAN) traffic. However, realizing its full potential faces multiple challenges. First, due to containers’ fine-grained computation resource isolation, finely configuring services’ computation resource requirements is needed, especially for resource-constrained edge nodes. Second, due to edges’ heterogeneity and services’ diversity, system performance highly depends on which edge node to place each service. Third, as the main metric of quality of service, service response time involves the computing-network delay tradeoff and urges to optimize the decisions jointly. Prior works on edge-enabled service placement either ignore computation resource isolation and configuration, or assume computation resource configuration is given manually. To fill this gap, this paper investigates the joint service placement and computation configuration problem for provisioning containerized services at edges. Then based on the convex and submodular optimization techniques, we propose a two-stage greedy and local-search combined algorithm, TeLa for short. Rigorous theoretical analyses demonstrate that TeLa is a polynomial-time algorithm with performance guarantees. Finally, we implement twelve containerized services and an edge computing prototype to realistically evaluate TeLa. The results confirm TeLa’s empirical superiority over state-of-the-art algorithms, in terms of 39% on average reduction on the weighted sum of service response time and WAN traffic.

Improving Scientific Literature Classification: A Parameter-Efficient Transformer-Based Approach

Transformer-based models have been utilized in natural language processing (NLP) for a wide variety of tasks like summarization, translation, and conversational agents. These models can capture long-term dependencies within the input, so they have significantly more representational capabilities than Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). Nevertheless, these models require significant computational resources in terms of high memory usage, and extensive training time. In this paper, we propose a novel document categorization model, with improved parameter efficiency that encodes text using a single, lightweight, multiheaded attention encoder block. The model also uses a hybrid word and position embedding to represent input tokens. The proposed model is evaluated for the Scientific Literature Classification task (SLC) and is compared with state-of-the-art models that have previously been applied to the task. Ten datasets of varying sizes and class distributions have been employed in the experiments. The proposed model shows significant performance improvements, with a high level of efficiency in terms of parameter and computation resource requirements as compared to other transformer-based models, and outperforms previously used methods.

Seamless Wireless Communication Platform for Internet of Things Applications

The rapid growth of the Internet of Things (IoT) devices resulted in the proliferation of wireless technologies to cater to their increasing data rate requirements and support multiple applications. However, such ever-increasing wireless technologies present numerous challenges such as incompatible wireless standards, increased energy consumption, and insecure communication. The traditional gateways proposed in the literature suffers from limitations such as computational complexity, resource requirements, increased cost, and device size. We vision an era of seamless wireless communication to alleviate the aforementioned challenges in IoT applications. through three inter-dependent functionalities namely detection and identification of wireless technologies, energy-efficient transmit power control, and secure end-to-end communication. To prove the concept, a new gateway is proposed to achieve these three functionalities with only physical layer measurements so that the different communication protocols in the higher layers can be avoided. Novel schemes are conceptualized for resource-limited seamless IoT applications. Moreover, the conceptual seamless IoT platform is validated through software-based computer simulation and software-defined radio-based testbed implementation. The preliminary analysis demonstrates that the proposed platform has great potential in advancing seamless IoT applications.

A High‐Resolution Prediction Network for Predicting Intratumoral Distribution of Nanoprobes by Tumor Vascular and Nuclear Feature

In this study, the critical need for precise and accurate prediction of intra‐tumor heterogeneity related to the enhanced permeability and retention effect and spatial distribution of nanoprobes is addressed for the development of effective nanodrug delivery strategies. Current predictive models are limited in terms of resolution and accuracy, prompting the construction of a high‐resolution prediction network (HRPN) that estimates the microdistribution of quantum dots, factoring in tumor vascular and nuclear features. The HRPN algorithm is trained using 27 780 patches and validated on 4920 patches derived from 4T1 breast cancer whole‐slide images, demonstrating its reliability. The HRPN model exhibits minimal error (mean square error = 1.434, root mean square error = 1.198), satisfactory goodness of fit (R2 = 0.891), and superior image quality (peak signal‐to‐noise ratio = 44.548) when compared to a generative‐adversarial‐network‐structured model. Furthermore, the HRPN model offers improved prediction accuracy, broader prediction intervals, and reduced computational resource requirements. Consequently, the proposed model yields high‐resolution predictions that more closely resemble actual tumor microdistributions, potentially serving as a powerful analytical tool for investigating the spatial relationship between the tumor microenvironment and nanoprobes.

Inertia matrix identification of combined spacecraft using a deep neural network with optimized network structure

After the space robot captures the out-of-control spacecraft, they form a combined spacecraft. For the parameter identification problem of combined spacecraft with complex angular rate measurement noise, deep learning could provide a new solution, due to its ability to extract robust features from noisy data. However, the deep neural network (DNN) has substantial computing resource requirements, which restricts its application in spacecraft systems. To solve the above problems, an inertia matrix identification approach combining a structure optimization strategy of DNN is proposed to identify the inertia matrix in the presence of complex measurement noise using the DNN with optimized structure. The attitude dynamics of combined spacecraft are firstly derived, and the identification problem considering complex measurement noise is described. Based on the dynamics model, the construction process of DNN for accurately identifying the inertia matrix is proposed. A structure optimization strategy of DNN is then designed to optimize the structure of DNN for increasing computing efficiency. This strategy provides a whole set of solutions including the structure selection of the initial DNN, the redundant neurons pruning process, and the end-of-optimization condition. By combining the proposed construction process of DNN, the structure of the DNN is finally optimized by iterating the structured pruning and retraining step. The performance of the proposed DNN with optimized network structure is verified on a testing set with complex measurement noise. The results show that the proposed approach can accurately identify the inertia matrix of combined spacecraft, and meanwhile greatly improve the computing efficiency of DNN.

Digital Twin of Intelligent Small Surface Defect Detection with Cyber-manufacturing Systems

With the remarkable technological development in cyber-physical systems, industry 4.0 has evolved by use of a significant concept named digital twin (DT). However, it is still difficult to construct a relationship between twin simulation and a real scenario considering dynamic variations, especially when dealing with small surface defect detection tasks with high performance and computation resource requirements. In this article, we aim to construct cyber-manufacturing systems to achieve a DT solution for small surface defect detection task. Focusing on DT-based solution, the proposed system consists of an Edge–Cloud architecture and a surface defect detection algorithm. Considering dynamic characteristics and real-time response requirement, Edge–Cloud architecture is built to achieve smart manufacturing by efficiently collecting, processing, analyzing, and storing data produced by factory. A deep learning–based algorithm is then constructed to detect surface defeats based on multi-modal data, i.e., imaging and depth data. Experiments show the proposed algorithm could achieve high accuracy and recall in small defeat detection task, thus constructing DT in cyber-manufacturing.

TARS: A parallel tetrahedral discontinuous finite element code for the solution of the discrete ordinates neutron transport equation

In this study, we introduce TARS, a neutron transport code based on the discrete-ordinates discontinuous finite element (SN-DFEM) method, which tackles existing challenges in the field by providing accurate and efficient solutions to the neutron transport equation. The use of DFEM enables higher-order spatial discretization, improving numerical stability, while parallelizing transport sweeping reduces computational time and resource requirements. TARS' performance was evaluated using the Takeda benchmark problems. The results showed that TARS generates highly accurate solutions, with eigenvalue errors no more than 90 pcm compared to MC solutions. The relative differences of region-averaged fluxes were below 1% in most regions, except those with severe flux gradients. Parallel calculation efficiency analysis revealed a PCE of over 70% even at 1024 CPU cores. However, hybrid MPI and OpenMP parallel programming did not outperform pure MPI programming, suggesting that the latter might be more suitable for TARS.

Automated models of visual information processing

The article presents a study devoted to the development and research of an automated model of visual information processing. The goal of the research was to create a comprehen-sive model capable of automatically processing and analyzing various forms of visual data, such as images and videos. The model is developed on the basis of a combined approach that combines various algorithms and methods of visual information processing. The literature review conducted within the scope of this study allowed us to study the existing methods and algorithms for visual information processing. Various image processing approaches were analyzed, including segmentation, pattern recognition, object classification and detection, video analysis, and other aspects. As a result of the review, the advantages and limitations of each approach were identified, as well as the areas of their application were determined. The developed model showed high accuracy and efficiency in visual data processing. It can suc-cessfully cope with the tasks of segmentation, recognition and classification of objects, as well as video analysis. The results of the study confirmed the superiority of the proposed model. Potential applications of the automated model are considered, such as medicine, robotics, security, and many others. However, limitations of the model such as computational resource requirements and quality of input data are also noted. Further development of this research can be aimed at optimizing the model, adapting it to specific tasks and expanding its func-tionality. In general, the study confirms the importance of automated models of visual infor-mation processing and its important place in modern technologies. The results of the research can be useful for the development of new systems based on visual data processing and con-tribute to progress in the field of computer vision and artificial intelligence.

Reduced-Dimension Surrogate Modeling to Characterize the Damage Tolerance of Composite/Metal Structures

Complex engineering models are typically computationally demanding and defined by a high-dimensional parameter space challenging the comprehensive exploration of parameter effects and design optimization. To overcome this curse of dimensionality and to minimize computational resource requirements, this research demonstrates a user-friendly approach to formulating a reduced-dimension surrogate model that represents a high-dimensional, high-fidelity source model. This approach was developed specifically for a non-expert using commercially available tools. In this approach, the complex physical behavior of the high-fidelity source model is separated into individual, interacting physical behaviors. A separate reduced-dimension surrogate model is created for each behavior and then all are summed to formulate the reduced-dimension surrogate model representing the source model. In addition to a substantial reduction in computational resources and comparable accuracy, this method also provides a characterization of each individual behavior providing additional insight into the source model behavior. The approach encompasses experimental testing, finite element analysis, surrogate modeling, and sensitivity analysis and is demonstrated by formulating a reduced-dimension surrogate model for the damage tolerance of an aluminum plate reinforced with a co-cured bonded E-glass/epoxy composite laminate under four-point bending. It is concluded that this problem is difficult to characterize and breaking the problem into interacting mechanisms leads to improved information on influential parameters and efficient reduced-dimension surrogate modeling. The disbond damage at the interface between the resin and metal proved the most difficult mechanism for reduced-dimension surrogate modeling as it is only engaged in a small subspace of the full parameter space. A binary function was successful in engaging this damage mechanism when applicable based on the values of the most influential parameters.

Extracting laboratory test information from paper-based reports

BackgroundIn the healthcare domain today, despite the substantial adoption of electronic health information systems, a significant proportion of medical reports still exist in paper-based formats. As a result, there is a significant demand for the digitization of information from these paper-based reports. However, the digitization of paper-based laboratory reports into a structured data format can be challenging due to their non-standard layouts, which includes various data types such as text, numeric values, reference ranges, and units. Therefore, it is crucial to develop a highly scalable and lightweight technique that can effectively identify and extract information from laboratory test reports and convert them into a structured data format for downstream tasks.MethodsWe developed an end-to-end Natural Language Processing (NLP)-based pipeline for extracting information from paper-based laboratory test reports. Our pipeline consists of two main modules: an optical character recognition (OCR) module and an information extraction (IE) module. The OCR module is applied to locate and identify text from scanned laboratory test reports using state-of-the-art OCR algorithms. The IE module is then used to extract meaningful information from the OCR results to form digitalized tables of the test reports. The IE module consists of five sub-modules, which are time detection, headline position, line normalization, Named Entity Recognition (NER) with a Conditional Random Fields (CRF)-based method, and step detection for multi-column. Finally, we evaluated the performance of the proposed pipeline on 153 laboratory test reports collected from Peking University First Hospital (PKU1).ResultsIn the OCR module, we evaluate the accuracy of text detection and recognition results at three different levels and achieved an averaged accuracy of 0.93. In the IE module, we extracted four laboratory test entities, including test item name, test result, test unit, and reference value range. The overall F1 score is 0.86 on the 153 laboratory test reports collected from PKU1. With a single CPU, the average inference time of each report is only 0.78 s.ConclusionIn this study, we developed a practical lightweight pipeline to digitalize and extract information from paper-based laboratory test reports in diverse types and with different layouts that can be adopted in real clinical environments with the lowest possible computing resources requirements. The high evaluation performance on the real-world hospital dataset validated the feasibility of the proposed pipeline.

Transfer learning from T1‐weighted to T2‐weighted Magnetic resonance sequences for brain image segmentation

AbstractMagnetic resonance (MR) imaging is a widely employed medical imaging technique that produces detailed anatomical images of the human body. The segmentation of MR images plays a crucial role in medical image analysis, as it enables accurate diagnosis, treatment planning, and monitoring of various diseases and conditions. Due to the lack of sufficient medical images, it is challenging to achieve an accurate segmentation, especially with the application of deep learning networks. The aim of this work is to study transfer learning from T1‐weighted (T1‐w) to T2‐weighted (T2‐w) MR sequences to enhance bone segmentation with minimal required computation resources. With the use of an excitation‐based convolutional neural networks, four transfer learning mechanisms are proposed: transfer learning without fine tuning, open fine tuning, conservative fine tuning, and hybrid transfer learning. Moreover, a multi‐parametric segmentation model is proposed using T2‐w MR as an intensity‐based augmentation technique. The novelty of this work emerges in the hybrid transfer learning approach that overcomes the overfitting issue and preserves the features of both modalities with minimal computation time and resources. The segmentation results are evaluated using 14 clinical 3D brain MR and CT images. The results reveal that hybrid transfer learning is superior for bone segmentation in terms of performance and computation time with DSCs of 0.5393 ± 0.0007. Although T2‐w‐based augmentation has no significant impact on the performance of T1‐w MR segmentation, it helps in improving T2‐w MR segmentation and developing a multi‐sequences segmentation model.

A Homomorphic Encryption Framework for Privacy-Preserving Spiking Neural Networks

Machine learning (ML) is widely used today, especially through deep neural networks (DNNs); however, increasing computational load and resource requirements have led to cloud-based solutions. To address this problem, a new generation of networks has emerged called spiking neural networks (SNNs), which mimic the behavior of the human brain to improve efficiency and reduce energy consumption. These networks often process large amounts of sensitive information, such as confidential data, and thus privacy issues arise. Homomorphic encryption (HE) offers a solution, allowing calculations to be performed on encrypted data without decrypting them. This research compares traditional DNNs and SNNs using the Brakerski/Fan-Vercauteren (BFV) encryption scheme. The LeNet-5 and AlexNet models, widely-used convolutional architectures, are used for both DNN and SNN models based on their respective architectures, and the networks are trained and compared using the FashionMNIST dataset. The results show that SNNs using HE achieve up to 40% higher accuracy than DNNs for low values of the plaintext modulus t, although their execution time is longer due to their time-coding nature with multiple time steps.

A Unified Parallel CORDIC-Based Hardware Architecture for LSTM Network Acceleration

Deep Neural Networks (DNNs) have recently become the standard tool for solving various practical problems in a wide range of applications with state-of-the-art performance. Recurrent Neural Networks (RNNs) such as Long Short-Term Memory (LSTM) are a subset of DNNs with fully connected single or multi-layer networks. The complex neurons and internal states of LSTM networks enable them to build a memory of events, making them ideal for time series applications. Despite the great potential of LSTM networks, their heterogeneous operations and computational resource requirements create a vast gap when it comes to the fast processing time required in real-time applications using low-power, low-cost edge devices. This work proposes a novel hardware architecture that combines serial-parallel computation with matrix algebra concepts and efficient low-power computer arithmetics for LSTM network acceleration. The hardware is based on a systolic ring of outer-product-based processing elements (PEs) and a reusable single activation function block (AFB). PEs and AFB are implemented using the coordinate rotation digital computer algorithm (CORDIC) in the linear and hyperbolic modes. Unlike most approaches, the proposed hardware can be configured to perform recurrent and non-recurrent fully connected layers (FC) computations, making it suitable for various low-power edge applications. The architecture is validated on the Xilinx PYNQ-Z1 development board using an open-source time series dataset. The implemented design achieves <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$114 \mu s$</tex-math></inline-formula> average latency and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$1.8GOPS$</tex-math></inline-formula> throughput. The proposed design's low latency and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$0.438W$</tex-math></inline-formula> power consumption makes it suitable for resource-constrained edge platforms.

Model of elemental data flow distribution in the internet of things supporting fog platform

The subject of the research is models and methods for optimizing resource and task management in the fog computing environment of the Internet of Things (IoT). The increasing number of connected devices and the volumes of data collected in IoT networks make it essential to improve management systems that ensure the optimal distribution of tasks and resources. Fog computing addresses this challenge by distributing computational tasks closer to data sources and end-users. The goal of this work is to enhance the efficiency of fog computing technologies to achieve optimal task and resource allocation in IoT networks. The main tasks of this work are as follows. Firstly, considering the diverse requirements of computational resources and tasks in IoT, reviewing existing methods and developments in the field is necessary. Secondly, it is essential to investigate and compare clustering methods, particularly DBSCAN and C-Means, for effective resource management. The DBSCAN clustering method enables efficient task distribution based on their location, while the C-Means method allows grouping resources based on their characteristics. The final task involves developing a mathematical model that considers input parameters such as system response, cluster resource requirements, data proximity to processing, etc. This model will enable the analysis of potential scenarios and decision-making regarding the optimal distribution of tasks and resources in the IoT environment. Conclusion. This research aims to solve the urgent problem of managing resources and tasks in the fog IoT environment. A review of existing methods and developments in resource and task management in IoT is conducted. DBSCAN and C-Means clustering methods are compared to determine their effectiveness in resource management. A set-theoretic model is developed that considers various parameters for making optimal decisions on the distribution of tasks and resources. It is established that the use of clustering methods and the developed model help to improve system performance and ensure more efficient use of fog computing resources in the IoT environment.

HivNet: Studying in Depth the Morphology of HIV-1 Virion Using Deep Learning

Purpose: Human Immunodeficiency Virus (HIV) continues to be a disease that kills thousands of individuals each year. The HIV infection is incurable. However, HIV infection has turned into a treatable chronic health condition because of improved access to efficient HIV prevention, diagnosis, treatment, and care. Transmission Electron Microscopy’s (TEM) ability to directly visualize virus particles and distinguish ultrastructure morphology at the nanometer scale, makes it useful in HIV-1 research where it is used for assessing the actions of inhibitors that obstruct the maturation and morphogenesis phases of the virus lifecycle. Hence with its use, the disease's serious stage can be avoided by receiving an early diagnosis.&#x0D; Materials and Methods: Through the dedicated use of computer vision frameworks and machine learning techniques, we have developed an optimized low-computational-cost 8-layer Convolutional Neural Network (CNN) backbone capable of classifying HIV-1 virions at various stages of maturity and morphogenesis. The dataset including TEM images of HIV-1 viral life cycle phases is analysed and augmented through various techniques to make the framework robust in real-time. The CNN layers then extract pertinent disease traits from TEM images and utilise them to provide diagnostic predictions.&#x0D; Results: It was discovered that the framework performed with an accuracy of 99.76% on the training set, 85.83% on the validation set, and 91.33% on the test set, after being trained on a wide range of micrographs which comprised of different experimental samples and magnifications.&#x0D; Conclusion: The suggested network's performance was compared to that of other state-of-the-art networks, and it was discovered that the proposed model was undisputed for classifying TEM images of unseen HIV-1 virion and required less time to train and tweak its weights. The framework can operate more effectively than machine learning algorithms that consume a lot of resources and can be deployed with limited computation and memory resource requirements.

Modelling and simulation of surface formation in electrical discharge machining based on thermo-hydraulic coupling

Surface formation in electrical discharge machining (EDM) involves melting, evaporation, re-solidification and crater overlapping. However, almost all existing models concerning surface formation in EDM only consider the thermal process and ignore hydrodynamic phenomena during machining. Herein, a novel three-dimensional thermo-hydraulic coupling model incorporating the arbitrary Lagrangian–Eulerian scheme was developed to address the above problem. The model comprehensively considered the surface deformation caused by melt flow, temperature-dependent material properties, phase changes and the effects of practical forces on melt pools to simulate surface formation on an anodic workpiece at a discharge current of 2 A. Simulation results demonstrate that during single-pulse discharge, molten material flowed radially to the crater periphery. The radial melt flow was driven by evaporation recoil pressure and thermocapillary force. The effect of evaporation recoil pressure on melt flow was gradually counteracted by that of surface tension after discharge ignition, and thermocapillary force became the dominant factor that maintained the existence of bulges around craters. During multiple discharges, the bulges generated by different discharges underwent superposition owing to the flow of molten material and subsequently formed surface peaks. The simulation results also provide the measurements of crater diameter, surface roughness and recast layer thickness, which are consistent with experimental results. The proposed modelling method can provide a useful reference for further investigations on surface integrity in EDM with minimal computing resource requirements.

Model Compression for Communication Efficient Federated Learning.

Despite the many advantages of using deep neural networks over shallow networks in various machine learning tasks, their effectiveness is compromised in a federated learning setting due to large storage sizes and high computational resource requirements for training. A large model size can potentially require infeasible amounts of data to be transmitted between the server and clients for training. To address these issues, we investigate the traditional and novel compression techniques to construct sparse models from dense networks whose storage and bandwidth requirements are significantly lower. We do this by separately considering compression techniques for the server model to address downstream communication and the client models to address upstream communication. Both of these play a crucial role in developing and maintaining sparsity across communication cycles. We empirically demonstrate the efficacy of the proposed schemes by testing their performance on standard datasets and verify that they outperform various state-of-the-art baseline schemes in terms of accuracy and communication volume.

Estimation of the impact of combined sewer overflows on surface water quality in a sparsely monitored area

Combined sewer overflows (CSOs) can have a severe negative, local impact on surface water systems. To assure good ecological surface water quality and drinking water production that meets the demands, the impact of sewer system overflows on the surrounding water bodies for current and future climate conditions needs to be assessed. Typically, integrated, detailed hydrological and hydrodynamic water quantity and quality models are used for this purpose, but often data and computational resource requirements limit their applicability. Therefore, an alternative computationally efficient, integrated water quantity and quality model of sewer systems and their receiving surface waters is proposed to assess the impact of CSOs on surface water quality in a sparsely observed area. A conceptual model approach to estimate CSO discharges is combined with an empirical model for estimating CSO pollutant concentrations based on waste water treatment plant influent observations. Both methods are compared with observations and independent results of established reference methods as to evaluate their performance. The methodology is demonstrated by modelling the current impact of CSOs on the water abstraction area of a major drinking water production centre in Flanders, Belgium. It is concluded that the proposed conceptual models achieve similar results for daily WWTP effluent and CSO frequency, whereby the accumulated CSO volume is similar to more detailed full hydrodynamic models. Further, the estimated pollutant concentrations correspond with another dataset based on high resolution sampled overflows. As a result, the proposed computational efficient method can give insights in the impact of CSOs on the water quality at a catchment level and can be used for planning monitoring campaigns or performing analyses of e.g. the current and future water availability for a data scarce areas.

Impact of H&E Stain Normalization on Deep Learning Models in Cancer Image Classification: Performance, Complexity, and Trade-Offs.

Accurate classification of cancer images plays a crucial role in diagnosis and treatment planning. Deep learning (DL) models have shown promise in achieving high accuracy, but their performance can be influenced by variations in Hematoxylin and Eosin (H&E) staining techniques. In this study, we investigate the impact of H&E stain normalization on the performance of DL models in cancer image classification. We evaluate the performance of VGG19, VGG16, ResNet50, MobileNet, Xception, and InceptionV3 on a dataset of H&E-stained cancer images. Our findings reveal that while VGG16 exhibits strong performance, VGG19 and ResNet50 demonstrate limitations in this context. Notably, stain normalization techniques significantly improve the performance of less complex models such as MobileNet and Xception. These models emerge as competitive alternatives with lower computational complexity and resource requirements and high computational efficiency. The results highlight the importance of optimizing less complex models through stain normalization to achieve accurate and reliable cancer image classification. This research holds tremendous potential for advancing the development of computationally efficient cancer classification systems, ultimately benefiting cancer diagnosis and treatment.

Stability and Equilibrium Structures of Unknown Ternary Metal Oxides Explored by Machine-Learned Potentials.

Ternary metal oxides are crucial components in a wide range of applications and have been extensively cataloged in experimental materials databases. However, there still exist cation combinations with unknown stability and structures of their compounds in oxide forms. In this study, we employ extensive crystal structure prediction methods, accelerated by machine-learned potentials, to investigate these untapped chemical spaces. We examine 181 ternary metal oxide systems, encompassing most cations except for partially filled 3d or f shells, and determine their lowest-energy crystal structures with representative stoichiometry derived from prevalent oxidation states or recommender systems. Consequently, we discover 45 ternary oxide systems containing stable compounds against decomposition into binary or elemental phases, the majority of which incorporate noble metals. Comparisons with other theoretical databases highlight the strengths and limitations of informatics-based material searches. With a relatively modest computational resource requirement, we contend that heuristic-based structure searches, as demonstrated in this study, offer a promising approach for future materials discovery endeavors.