Resource Requirements Research Articles

On the use of synthetic data for body detection in maritime search and rescue operations

Time is a critical factor in maritime Search And Rescue (SAR) missions, during which promptly locating survivors is paramount. Unmanned Aerial Vehicles (UAVs) are a useful tool with which to increase the success rate by rapidly identifying targets. While this task can be performed using other means, such as helicopters, the cost-effectiveness of UAVs makes them an effective choice. Moreover, these vehicles allow the easy integration of automatic systems that can be used to assist in the search process. Despite the impact of artificial intelligence on autonomous technology, there are still two major drawbacks to overcome: the need for sufficient training data to cover the wide variability of scenes that a UAV may encounter and the strong dependence of the generated models on the specific characteristics of the training samples. In this work, we address these challenges by proposing a novel approach that leverages computer-generated synthetic data alongside novel modifications to the You Only Look Once (YOLO) architecture that enhance its robustness, adaptability to new environments, and accuracy in detecting small targets. Our method introduces a new patch-sample extraction technique and task-specific data augmentation, ensuring robust performance across diverse weather conditions. The results demonstrate our proposal’s superiority, showing an average 28% relative improvement in mean Average Precision (mAP) over the best-performing state-of-the-art baseline under training conditions with sufficient real data, and a remarkable 218% improvement when real data is limited. The proposal also presents a favorable balance between efficiency, effectiveness, and resource requirements.

Advanced Image Preprocessing and Integrated Modeling for UAV Plant Image Classification

The automatic identification of plant species using unmanned aerial vehicles (UAVs) is a valuable tool for ecological research. However, challenges such as reduced spatial resolution due to high-altitude operations, image degradation from camera optics and sensor limitations, and information loss caused by terrain shadows hinder the accurate classification of plant species from UAV imagery. This study addresses these issues by proposing a novel image preprocessing pipeline and evaluating its impact on model performance. Our approach improves image quality through a multi-step pipeline that includes Enhanced Super-Resolution Generative Adversarial Networks (ESRGAN) for resolution enhancement, Contrast-Limited Adaptive Histogram Equalization (CLAHE) for contrast improvement, and white balance adjustments for accurate color representation. These preprocessing steps ensure high-quality input data, leading to better model performance. For feature extraction and classification, we employ a pre-trained VGG-16 deep convolutional neural network, followed by machine learning classifiers, including Support Vector Machine (SVM), random forest (RF), and Extreme Gradient Boosting (XGBoost). This hybrid approach, combining deep learning for feature extraction with machine learning for classification, not only enhances classification accuracy but also reduces computational resource requirements compared to relying solely on deep learning models. Notably, the VGG-16 + SVM model achieved an outstanding accuracy of 97.88% on a dataset preprocessed with ESRGAN and white balance adjustments, with a precision of 97.9%, a recall of 97.8%, and an F1 score of 0.978. Through a comprehensive comparative study, we demonstrate that the proposed framework, utilizing VGG-16 for feature extraction, SVM for classification, and preprocessed images with ESRGAN and white balance adjustments, achieves superior performance in plant species identification from UAV imagery.

Thermodynamic linear algebra

AbstractLinear algebra is central to many algorithms in engineering, science, and machine learning; hence, accelerating it would have tremendous economic impact. Quantum computing has been proposed for this purpose, although the resource requirements are far beyond current technological capabilities. We consider an alternative physics-based computing paradigm based on classical thermodynamics, to provide a near-term approach to accelerating linear algebra. At first sight, thermodynamics and linear algebra seem to be unrelated fields. Here, we connect solving linear algebra problems to sampling from the thermodynamic equilibrium distribution of a system of coupled harmonic oscillators. We present simple thermodynamic algorithms for solving linear systems of equations, computing matrix inverses, and computing matrix determinants. Under reasonable assumptions, we rigorously establish asymptotic speedups for our algorithms, relative to digital methods, that scale linearly in matrix dimension. Our algorithms exploit thermodynamic principles like ergodicity, entropy, and equilibration, highlighting the deep connection between these two seemingly distinct fields, and opening up algebraic applications for thermodynamic computers.

Efficient I/O Performance-Focused Scheduling in High-Performance Computing

High-performance computing (HPC) systems are becoming increasingly important as contemporary exascale applications with demand extensive computational and data processing capability. To optimize these systems, efficient scheduling of HPC applications is important. In particular, because I/O is a shared resource among applications and is becoming more important due to the emergence of big data, it is possible to improve performance by considering the architecture of HPC systems and scheduling jobs based on I/O resource requirements. In this paper, we propose a scheduling scheme that prioritizes HPC applications based on their I/O requirements. To accomplish this, our scheme analyzes the IOPS of scheduled applications by examining their execution history. Then, it schedules the applications at pre-configured intervals based on their expected IOPS to maximize the available IOPS across the entire system. Compared to the existing first-come first-served (FCFS) algorithm, experimental results using real-world HPC log data show that our scheme reduces total execution time by 305 h and decreases costs by USD 53 when scheduling 10,000 jobs utilizing public cloud resources.

Coupling Coordination and Data Management for Additive Manufacturing Systems Based on Smart Logistics

Additive Manufacturing (AM) has rapidly advanced due to its benefits in reducing material consumption and resource requirements, offering significant opportunities for energy-efficient and environmentally friendly production. As manufacturing technologies evolve, there is growing interest in smart manufacturing, prompting both academia and industry to explore ways to enhance efficiency and sustainability. Effective logistics management plays a crucial role in this context, requiring precise data prediction and optimization. This article introduces a Smart Logistics service to foster dynamic coordination within the smart logistics ecosystem. The first step is to integrate smart logistics components that can communicate effectively and provide specific value-added services. This integration aims to support contemporary industrial and production environments by enhancing decision-making models. The second step involves applying the Coupling Coordination model to assess the development of logistics in conjunction with broader economic growth. This model helps in constructing a high-quality development index system for logistics, which is essential for evaluating the efficiency of logistics operations. The third step is to evaluate the coordination between the automotive, digitalization, and logistics sectors. This assessment provides insights into the overall development levels of cities and highlights the significant impact on sustainable growth within the automotive industry. The final step includes analyzing the progress of coupling coordination in Indian provinces, identifying the transition from minor imbalances to primary coordination, and forecasting that strong coordination will be achieved by 2025. This analysis underscores the potential for mutual growth and cooperation by elevating the logistics sector’s standards and strengthening inter-industry relationships.

Comparison of Controlled-Z operation and beam-splitter transformation for generation of squeezed Fock states by measurement

In the present paper, we investigate the protocol for the generation of squeezed Fock states by measuring the number of particles from a two-mode entangled Gaussian state using a beam splitter and a controlled-Z operation. From two different perspectives, we analyzed two entanglement transformations in the protocol. We evaluated the energy costs and resource requirements of the analyzed schemes. Furthermore, we studied the impact of experimental imperfections on the non-Gaussian states generated by measuring the number of particles. We explored the effects of photon loss and imperfect detectors on the squeezed Fock state generation protocol.

A Review on Multiple-Ocular Disease Detection Methodology using ML and DL Techniques

The rapid advancement in machine learning (ML) and deep learning (DL) techniques has significantly impacted the detection and diagnosis of ocular diseases, which are critical for preserving vision and overall eye health. This review aims to explore the various ML and DL methodologies applied to the detection of multiple ocular diseases, highlighting their effectiveness, limitations, and areas for improvement. The motivation behind this review stems from the increasing prevalence of ocular diseases and the need for efficient, accurate diagnostic tools. Despite the promising results of existing techniques, limitations such as data variability, the need for extensive training data, and computational resource requirements persist. The objective is to synthesize current methodologies and propose enhancements, particularly through the integration of attention mechanisms in convolutional neural networks (CNNs). This review identifies gaps in current research and suggests directions for future work to enhance diagnostic accuracy and clinical applicability.

Plant‐based protein crops and their improvement: Current status and future perspectives

AbstractThe plant‐based protein industry is rapidly growing and is projected to grow to over $27 billion by 2030. In addition to monetary benefits, plant‐based protein crops, particularly pulse crops (i.e., legume species with dry edible seed), can supplement the existing crop biodiversity and break the disease and insect pest cycle in drastically changing climatic conditions. Most commercially available plant‐based protein ingredients come from only 2% of the 150 plant species on which our food supply depends. Therefore, it is imperative to diversify the cropping system with pulse crops that provide a more nutritious, climate‐resilient, sustainable food value chain. At present, a large portion of US pulse crop production is exported, yet the shifting demographics, dietary preferences, and a rise in domestic consumption present enormous potential for existing and emerging pulse crops, as well as their breeding. Despite a rise in demand from consumers and industries, farmers are yet to fully reap the benefits of these crops due to a shortage of improved varieties. These new varieties could lead to a diverse, sustainable protein supply that satisfies the growing domestic and global demand for plant‐based foods. This review intends to comprehensively explore the geographical and cultural acceptance of plant‐based proteins, manufacturing techniques, protein products, anti‐nutritional factors, techniques and tools for protein estimation, high‐throughput estimation of seed protein, protein composition requirements in food, breeding strategies, crop improvement, and resources to improve plant‐based protein pulse crops.

Container-Based Electronic Control Unit Virtualisation: A Paradigm Shift Towards a Centralised Automotive E/E Architecture

The past 40 years have seen automotive Electronic Control Units (ECUs) move from being purely mechanical controlled to being primarily digital controlled. While the safety of passengers and efficiency of vehicles has seen significant improvements, rising ECU numbers have resulted in increased vehicle weight, greater demands placed on power, more complex hardware and software, ad hoc methods for updating software, and subsequent increases in costs for both vehicle manufacturers and consumers. To address these issues, the research presented in this paper proposes that virtualisation technologies be applied within automotive electrical/electronic (E/E) architecture. The proposed approach is evaluated by comprehensively studying the CPU and memory resource requirements to support container-based ECU automotive functions. This comprehensive performance evaluation reveals that lightweight container virtualisation has the potential to welcome a paradigm shift in E/E architecture, promoting consolidation and enhancing the architecture by facilitating power, weight, and cost savings. Container-based virtualisation will also enable efficient and robust online dynamic software updates throughout a vehicle’s lifetime.

Segmentation-Based Detection for Luffa Seedling Grading Using the Seg-FL Model

This study addresses the issue of inaccurate and error-prone grading judgments in luffa plug seedlings. A new Seg-FL seedling segmentation model is proposed as an extension of the YOLOv5s-Seg model. The small leaves of early-stage luffa seedlings are liable to be mistaken for impurities in the plug trays. To address this issue, cross-scale connections and weighted feature fusion are introduced in order to integrate feature information from different levels, thereby improving the recognition and segmentation accuracy of seedlings or details by refining the PANet structure. To address the ambiguity of seedling edge information during segmentation, an efficient channel attention module is incorporated to enhance the network’s focus on seedling edge information and suppress irrelevant features, thus sharpening the model’s focus on luffa seedlings. By optimizing the CIoU loss function, the calculation of overlapping areas, center point distances, and aspect ratios between predicted and ground truth boxes is preserved, thereby accelerating the convergence process and reducing the computational resource requirements on edge devices. The experimental results demonstrate that the proposed model attains a mean average precision of 97.03% on a self-compiled luffa plug seedling dataset, representing a 6.23 percentage point improvement over the original YOLOv5s-Seg. Furthermore, compared to the YOLACT++, FCN, and Mask R-CNN segmentation models, the improved model displays increases in mAP@0.5 of 12.93%, 13.73%, and 10.53%, respectively, and improvements in precision of 15.73%, 16.93%, and 13.33%, respectively. This research not only validates the viability of the enhanced model for luffa seedling grading but also provides tangible technical support for the automation of grading in agricultural production.

ROSHAMBO: Open-Source Molecular Alignment and 3D Similarity Scoring.

Efficient virtual screening techniques are critical in drug discovery for identifying potential drug candidates. We present an open-source package for molecular alignment and 3D similarity calculations optimized for large-scale virtual screening of small molecules. This work parallels widely used proprietary tools and offers an approach complementary to structure-based virtual screening. Our package employs the PAPER software for optimizing molecular alignments based on Gaussian volume overlaps. GPU acceleration is utilized to significantly reduce computational time and resource requirements. After obtaining the optimal alignments between the target and the query molecules, both shape and color (based on pharmacophore features) scores are computed to assess molecular similarity, with aligned molecules optionally being output in sdf format. The package was benchmarked using the DUDE-Z public data sets. Results demonstrated the package's near-state-of-the-art performance and robustness across multiple target classes, with speed that enables many routine ligand-based drug discovery workflows. As an open-source and freely available resource (github.com/molecularinformatics/roshambo) with both a convenient Python API and command line interface, our package also addresses the need for accessible and efficient virtual screening tools in drug discovery.

Predictive brain activity related to auditory information is associated with performance in speech comprehension tasks in noisy environments

IntroductionUnderstanding speech in noisy environments is challenging even for individuals with normal hearing, and it poses a significant challenge for those with hearing impairments or listening difficulties. There are limitations associated with the current methods of evaluating speech comprehension in such environments, especially in individuals with peripheral hearing impairments. According to the predictive coding model, speech comprehension is an active inference process that integrates sensory information through the interaction of bottom-up and top-down processing. Therefore, in this study, we aimed to examine the role of prediction in speech comprehension using an electrophysiological marker of anticipation: stimulus-preceding negativity (SPN).MethodsWe measured SPN amplitude in young adults with normal hearing during a time-estimation task with auditory feedback under both quiet and noisy conditions.ResultsThe results showed that SPN amplitude significantly increased in noisy environments. Moreover, individual differences in SPN amplitude correlated with performance in a speech-in-noise test.DiscussionThe increase in SPN amplitude was interpreted as reflecting the increased requirement for attentional resources for accurate prediction of speech information. These findings suggest that SPN could serve as a noninvasive neural marker for assessing individual differences in top-down processing involved in speech comprehension in noisy environments.

CYP3A5 pharmacogenetic testing for tacrolimus in pediatric heart transplant patients: a budget impact analysis.

Pharmacogenomic testing can optimize drug efficacy and minimize adverse effects. CYP3A5 polymorphisms affect the metabolism of tacrolimus. We sought to estimate the budget impact of preemptive pharmacogenomic testing for CYP3A5 in pediatric heart transplantation patients from an institutional perspective. A decision tree was constructed to estimate the budget impact of pediatric heart transplant patients (age ≤18 years) initiated on tacrolimus with and without CYP3A5 pharmacogenomic testing. The budget impact of preemptive pharmacogenomic testing versus no pharmacogenomic testing was calculated. One-way sensitivity analysis and alternative analyses were conducted to assess the robustness of results to changes in model parameters. CYP3A5 genotype-guided dosing provided savings of up to $17 225 per patient compared to standard dosing. These savings decreased to $11 759 when using another institution's data for the standard-dosing group. The time to achieve therapeutic concentration in the poor metabolizer genotype-guided dosing group had the largest impact on cost savings while the cost of the pharmacogenetic test had the smallest impact on cost savings. Implementing CYP3A5 testing could save $17 225 per pediatric heart transplant patient receiving tacrolimus. As pharmacogenomic testing becomes more widespread, institutions should track resource requirements and outcomes to determine the best implementation policies going forward.

Start Time Planning for Cyclic Queuing and Forwarding in Time-Sensitive Networks

Time-sensitive networking (TSN) is a kind of network communication technology applied in fields such as industrial internet and intelligent transportation, capable of meeting the application requirements for precise time synchronization and low-latency deterministic forwarding. In TSN, cyclic queuing and forwarding (CQF) is a traffic shaping mechanism that has been extensively discussed in the recent literature, which allows the delay of time-triggered (TT) flow to be definite and easily calculable. In this paper, two algorithms are designed to tackle the start time planning issue with the CQF mechanism, namely the flow–path–offset joint scheduling (FPOJS) algorithm and congestion-aware scheduling algorithm, to improve the scheduling success ratio of TT flows. The FPOJS algorithm, which adopts a novel scheduling object—a combination of flow, path, and offset—implements scheduling in descending order of a well-designed priority that considers the resource capacity and resource requirements of ports. The congestion-aware scheduling algorithm identifies and optimizes congested ports during scheduling and substantially improves the scheduling success ratio by dynamically configuring port resources. The experimental results demonstrate that the FPOJS algorithm achieves a 39% improvement in the scheduling success ratio over the naive algorithm, 13% over the Tabu-ITP algorithm, and 10% over the MSS algorithm. Moreover, the algorithm exhibits a higher scheduling success ratio under large-scale TSN.

Bridging the Disparity: An Optimization Model for Equitable Allocation of Urban and Rural Educational Resources in Western China

The persistent disparity in educational resources between urban and rural areas in western China remains a critical barrier to achieving educational equity. This study addresses the critical issue by developing an innovative optimization model. The model aims to allocate financial, human, and physical resources effectively to minimize the educational quality gap between these regions. Using information from the China Education Panel Survey (CEPS) and concentrating on Xinjiang as a representative instance, our design integrates budget constraints, minimum resource requirements, and a unique regional prioritization factor. The goal of the objective function is to reduce the squared disparity in educational quality between city and rural regions. It thinks about variables such as Teacher-Student Ratio (TSR), Per Capita Investment (PCI), and Teaching Equipment Perfection (TEP). Our searchings for show that strategic resource distribution can considerably lessen educational inequalities in between metropolitan and country setups, especially by focusing on underdeveloped rural regions. Sensitivity analysis shows that customizing the local prioritization specification makes it possible for versatile policy strategies, ranging from equitable distribution to plainly sustaining rural regions. This study presents a pragmatic structure for boosting resource allotment in education. It supplies vital understandings for policymakers aiming to promote an extra equitable educational environment in Western China.

Tackling Pertussis in Schools: A Public Health Unit’s Approach

Abstract Issue Pertussis is a highly infectious bacterial respiratory disease, easily transmitted in closed settings, such as schools. In Portugal, 2019 was the last year with a significant number of cases (83). As pertussis usually follows epidemic cycles from 3 to 5 years, another wave is expected in 2024. Children aged 1-5 years are particularly vulnerable as they no longer benefit from maternal vaccination and have not yet received the 5-year boost recommended in the national schedule. Description of the problem In April 2024, the Cascais Public Health Unit (PHU) addressed two possible pertussis cases in preschools. A streamlined response protocol was drafted, including an online contact tracing form and task distribution among response team members, along with templates and a scripted guide for standardized communication. Results We were provided a list of close contacts for the children and staff and shared an online form. In the first case, 34 in 35 identified contacts responded to the form, while in the second case, 38 in 55. When comparing both scenarios, the average scripted call duration decreased from 10 to 6 minutes. The PHU response to the first index case took 6 hours, involving 4 physicians and 5 nurses, whereas to the second took 4 hours, involving 3 physicians and 3 nurses. However, implementation requires at least one team member with basic information technology skills to set up the system. Lessons Clear communication channels between schools and public health authorities are vital for effective outbreak management. Streamlining communication processes with online forms and scripted conversation guides improved efficiency. Centralised data collection reduced response time and resource requirements in each outbreak. Key messages • Established communication channel, standardized data collection, and streamlined protocol allow more effective outbreak management. • Centralized data collection via digital tools is particularly useful in limited human resources settings.

Implementation of the international health regulations – global challenges and the COVID-19 pandemic

Abstract The International Health Regulations (IHR) govern how 196 countries and the World Health Organization (WHO) collectively manage global disease spread while preserving international travel and commercial trade. The COVID-19 pandemic set an unprecedented challenge to the IHR commitments. To address this gap, two scoping reviews were conducted: one identifying global challenges in IHR implementation and the other analyzing challenges, successes and limitations of the IHR during the COVID-19 pandemic. The scoping reviews followed Arksey and O’Malley’s framework and PRISMA methodology, utilizing the PubMed, Medline, World of Science, SCOPUS, and Google Scholar databases, plus backwards citation chaining. Data selection and analysis were guided by pretest forms. From 128 articles screened, 52 met inclusion criteria for the IHR implementation review. 135 challenges were identified, in these categories: (1) requirement of financial and technical resources, (2) lack of enforcement and accountability mechanisms, (3) need for rules clarification, (4) deterrents to reporting, (5) inadequate governance, (6) insufficient cooperation between countries, (7) sovereignty issues, (8) scarcity of guidance and technical support from WHO, (9) lack of inclusion of roles for relevant non-state actors, (10) conflicts and emergencies. In the COVID-19 review, 15 articles were included. Although the role of the IHR during Health Emergencies was acknowledged, most countries faced limitations in its implementation during the pandemic. Key lessons included reinforcing the global health security strategy, reviewing the IHR to address current challenges and strengthening countries and states’ commitment to resolutions compliance. The ongoing debate over IHR updates confirms the need for its revision to safeguard global health. A more effective response grounded on the IHR is needed to face upcoming and expected events of COVID-19’s magnitude and severity.

A Multicenter Physician Survey Evaluating the Use of Ki-67 in Breast Cancer Management in Canada

Background: Ki-67’s response to pre-operative endocrine therapy (ET) in early breast cancer is an evidence-based tool to guide adjuvant treatment decisions. Physicians across Canada were surveyed to explore current practice patterns and perceived barriers to the use of Ki-67 in practice. Methods: Physicians were invited to participate in an anonymous survey and were eligible if they prescribed systemic therapy for breast cancer in Canada. Respondents were asked to describe their usage of Ki-67, perceptions of the evidence surrounding Ki-67 ET response, and interest in future trials using this approach. Results: The survey received 48/163 responses (29.4%). The majority of respondents (97.6%) reported access to Ki-67 testing upon request. Treatment decisions for adjuvant Abemaciclib was the most common reason (97.6%), followed by adjuvant chemotherapy decisions (16.7%). Only 19.0% had used Ki-67’s response to pre-operative ET in practice. Common barriers to this approach that were identified included a lack of awareness from other providers (54.8%), an increased resource requirement (54.8%), and a lack of timely medical oncology consultation (52.4%). The majority of physicians (85.3%) reported that they would participate in future trials using the Ki-67 endocrine response, and that rate of treatment decision change (95.2%) and cost analysis (42.3%) were important endpoints. Conclusions: Despite the widespread availability of Ki-67 testing, few physicians in Canada currently use it to assess endocrine response, predominantly due to logistical and resource constraints. There is a high level of interest in participating in future trials using this strategy, which should focus on disease related outcomes, feasibility, and the financial impact on the public healthcare system.

Improving ncRNA family prediction using Multi-Modal contrastive learning of sequence and structure.

Recent advancements in high-throughput sequencing technology have significantly increased the focus on non-coding RNA (ncRNA) research within the life sciences. Despite this, the functions of many ncRNAs remain poorly understood. Research suggests that ncRNAs within the same family typically share similar functions, underlining the importance of understanding their roles. There are two primary methods for predicting ncRNA families: biological and computational. Traditional biological methods are not suitable for large-scale data prediction due to the significant human and resource requirements. Concurrently, most existing computational methods either rely solely on ncRNA sequence data or are exclusively based on the secondary structure of ncRNA molecules. These methods fail to fully utilize the rich multimodal information available from ncRNAs, thereby preventing them from learning more comprehensive and in-depth feature representations. To tackle these problems, we proposed MM-ncRNAFP, a multi-modal contrastive learning framework for ncRNA family prediction. We first used a pre-trained language model to encode the primary sequences of a large mammalian ncRNA dataset. Then, we adopted a contrastive learning framework with an attention mechanism to fuse the secondary structure information obtained by graph neural networks. The MM-ncRNAFP method can effectively fuse multi-modal information. Experimental comparisons with several competitive baselines demonstrated that MM-ncRNAFP can achieve more comprehensive representations of ncRNA features by integrating both sequence and structural information. This integration significantly enhances the performance of ncRNA family prediction. Ablation experiments and qualitative analyses were performed to verify the effectiveness of each component in our model. Moreover, since our model is pre-trained on a large amount of ncRNA data, it has the potential to bring significant improvements to other ncRNA-related tasks. MM-ncRNAFP and the datasets are available at https://github.com/xuruiting2/MM-ncRNAFP.

The first kind of predictability problem of El Niño predictions in a multivariate coupled data‐driven model

AbstractEl Niño‐Southern Oscillation (ENSO) is the dominant atmosphere–ocean coupled mode of year‐to‐year variations in the tropical Pacific. It shows diverse spatiotemporal characteristics and casts major influences on seasonal predictions of global weather–climate extrema. Despite numerous dynamical and statistical models for ENSO prediction and predictability studies, they are commonly subjected to one‐to‐three issues among less skillful simulation of El Niño diversity, huge requirements of computational resources and a low robustness in statistics. Here, an efficient deep‐learning model involving nonlinear coupling of multiple variables is independently developed to study the predictability of two types of El Niño events related to initial uncertainty, which is the first kind of predictability problem. The model can skillfully simulate statistically robust features of observed El Niño diversity in terms of periodicity, amplitude, and seasonal phase‐locking. Using this model, we have revealed mathematically several new types of fastest‐growing initial errors in two types of El Niño predictions based on a novel concept of conditional nonlinear optimal perturbation (CNOP), especially including one that can strengthen central Pacific types of events, which is rarely investigated before. Moreover, CNOPs are superimposed into a numerical model, GFDL CM2p1, for comprehensive validation and growth mechanism mining, which demonstrates the consistent dynamical evolution of initial errors in both numerical and AI models. Our study represents the first attempt to explore the first kind of ENSO predictability problem from perspectives of nonlinear error‐evolving dynamics using a data‐driven model. This is of great importance as it offers us sufficient confidence to perform ENSO‐related (such as the Madden–Julian Oscillation, etc.) mechanisms and predictability studies in the future without strongly relying on dynamical numerical models.