Adaptive Metrics Research Articles

Locally led adaptation metrics for Africa: a framework for building resilience in smallholder farming sectors

Locally led adaptation metrics for Africa: a framework for building resilience in smallholder farming sectors

Community-Based Adaptation to Climate Change: Core Issues and Implications for Practical Implementations

According to current forecasts, global heating is likely to exceed 2.8 °C by the end of this century. This makes substantial adaptation measures necessary to secure a broad basis for livelihood provision and the conservation of biodiversity. While the implementation of top-down and technocratic adaptation efforts predominates, related adaptation shortcomings of a socio-economic and ecological nature are becoming more and more apparent. Community-based adaptation (CBA), with its participatory, inclusive and needs-based bottom-up approach, offers a promising and powerful alternative. This article uses a semi-systematic literature review approach to screen the current literature and identify core issues of CBA. Linking communality, locality, multidimensionality, power imbalances, transformative potential, localisation, the triad of adaptation metrics and nature-based adaptation to corresponding potential actions for practical implementations provides a more holistic conceptualisation and broadens the horizons for further learning, research and improved applications.

New Tree Routing Protocol with Adaptive Metrics Based on Hop Count

New Tree Routing Protocol with Adaptive Metrics Based on Hop Count

Semi-supervised adaptive anti-noise meta-learning for few-shot industrial gearbox fault diagnosis

Abstract Real-time and accurate predictive maintenance of industrial equipment is fundamental for ensuring the safety and stability of advanced manufacturing processes. Current fault diagnosis methods based on data mining rely on a large number of labeled samples, and obtaining sufficient labeled data for diagnosing industrial equipment faults is challenging. Meta-learning can achieve the diagnosis of few-shot samples to a certain extent, but the effect is not ideal. Semi-supervision can effectively leverage a large number of unlabeled samples, which is of great practical significance for handling scenarios involving limited labeled samples. However, noise interference can occur when unlabeled samples appear that do not belong to known categories. Therefore, this study proposes adaptive semi-supervised meta-learning networks for noisy few-shot gearbox fault diagnosis. Firstly, a residual network with a Morlet Wavelet layer is used to extract signal features. Next, sample-level attention is defined to select unlabeled samples that are more similar to labeled sample prototypes, thereby reducing the influence of noisy samples. The adaptive metric is used to obtain the relational distance functions of labeled samples and unlabeled samples. Adaptive semi-supervised meta-learning networks uses unlabeled data to refine prototypes for better fault diagnosis. The effectiveness and anti-noise performance of the proposed method are verified by using two gearbox datasets with various few-shot noise scenarios.

A Common Knowledge-Driven Generic Vision Inspection Framework for Adaptation to Multiple Scenarios, Tasks, and Objects.

The industrial manufacturing model is undergoing a transformation from a product-centric model to a customer-centric one. Driven by customized requirements, the complexity of products and the requirements for quality have increased, which pose a challenge to the applicability of traditional machine vision technology. Extensive research demonstrates the effectiveness of AI-based learning and image processing on specific objects or tasks, but few publications focus on the composite task of the integrated product, the traceability and improvability of methods, as well as the extraction and communication of knowledge between different scenarios or tasks. To address this problem, this paper proposes a common, knowledge-driven, generic vision inspection framework, targeted for standardizing product inspection into a process of information decoupling and adaptive metrics. Task-related object perception is planned into a multi-granularity and multi-pattern progressive alignment based on industry knowledge and structured tasks. Inspection is abstracted as a reconfigurable process of multi-sub-pattern space combination mapping and difference metric under appropriate high-level strategies and experiences. Finally, strategies for knowledge improvement and accumulation based on historical data are presented. The experiment demonstrates the process of generating a detection pipeline for complex products and continuously improving it through failure tracing and knowledge improvement. Compared to the (1.767°, 69.802 mm) and 0.883 obtained by state-of-the-art deep learning methods, the generated pipeline achieves a pose estimation ranging from (2.771°, 153.584 mm) to (1.034°, 52.308 mm) and a detection rate ranging from 0.462 to 0.927. Through verification of other imaging methods and industrial tasks, we prove that the key to adaptability lies in the mining of inherent commonalities of knowledge, multi-dimensional accumulation, and reapplication.

Interpretable metric learning in comparative metagenomics: The adaptive Haar-like distance.

Random forests have emerged as a promising tool in comparative metagenomics because they can predict environmental characteristics based on microbial composition in datasets where β-diversity metrics fall short of revealing meaningful relationships between samples. Nevertheless, despite this efficacy, they lack biological insight in tandem with their predictions, potentially hindering scientific advancement. To overcome this limitation, we leverage a geometric characterization of random forests to introduce a data-driven phylogenetic β-diversity metric, the adaptive Haar-like distance. This new metric assigns a weight to each internal node (i.e., split or bifurcation) of a reference phylogeny, indicating the relative importance of that node in discerning environmental samples based on their microbial composition. Alongside this, a weighted nearest-neighbors classifier, constructed using the adaptive metric, can be used as a proxy for the random forest while maintaining accuracy on par with that of the original forest and another state-of-the-art classifier, CoDaCoRe. As shown in datasets from diverse microbial environments, however, the new metric and classifier significantly enhance the biological interpretability and visualization of high-dimensional metagenomic samples.

Construction of an Adaptive Metric Model for Maximum Access Capacity of Distributed Photovoltaic in Active Distribution Networks

Distributed photovoltaic (PV) access to the active distribution network can easily cause power fluctuation and affect the reasonable setting. The maximum access capacity is constructed as an adaptive metric model. Taking the total installed capacity as the objective function, the maximum access capacity adaptive metering model is constructed. For the constructed model, the constraints of light abandonment constraint, energy storage system constraint, voltage deviation constraint at the common connection point, line limit capacity constraint and distribution transformer capacity constraint are set. The model is solved by using multiple swarm genetic algorithms, output maximum access capacity. The abandonment rate of PV nodes is less than 3%, and the voltage overrun rate and line capacity overrun rate are less than 1%.

Adaptation Metrics for Low-Order Nonstationary Line-of-Sight-MIMO Links

Adaptation Metrics for Low-Order Nonstationary Line-of-Sight-MIMO Links

A flexible approach to measure care coordination based on patient-sharing networks

BackgroundEffective care coordination may increase clinical efficiency, but its measurement remains difficult. The established metric “care density” (CD) measures care coordination based on patient-sharing among physicians, but it may be too rigid to generalize across disorders and countries. Therefore, we propose an extension called fragmented care density (FCD), which allows varying weights for connections between different types of providers. We compare both metrics in their ability to predict hospitalizations due to schizophrenia.MethodsWe conducted a longitudinal cohort study based on German claims data from 2014 through 2017 to predict quarterly hospital admissions. 21,016 patients with schizophrenia from the federal state Baden-Württemberg were included. CD and FCD were calculated based on patient-sharing networks. The weights of FCD were optimized to predict hospital admissions during the first year of a 24-month follow-up. Subsequently, we employed likelihood ratio tests to assess whether adding either CD or FCD improved a baseline model with control variables for the second follow-up year.ResultsThe inclusion of FCD significantly improved the baseline model, Χ2(1) = 53.30, p < 0.001. We found that patients with lower percentiles in FCD had an up to 21% lower hospitalization risk than those with median or higher values, whereas CD did not affect the risk.ConclusionsFCD is an adaptive metric that can weight provider relationships based on their relevance for predicting any outcome. We used it to better understand which medical specialties need to be involved to reduce hospitalization risk for patients with schizophrenia. As FCD can be modified for different health conditions and systems, it is broadly applicable and might help to identify barriers and promoting factors for effective collaboration.

Acceleration of Environmental Electrostatic Potential Using Cholesky Decomposition with Adaptive Metric (CDAM) for Fragment Molecular Orbital-based Molecular Dynamics (FMO-MD) Simulation

Acceleration of Environmental Electrostatic Potential Using Cholesky Decomposition with Adaptive Metric (CDAM) for Fragment Molecular Orbital-based Molecular Dynamics (FMO-MD) Simulation

Component Assessment of the Electric Transmission Grid to Hurricanes

AbstractThe increased frequency and intensity of extreme weather events from climate change necessitates understanding impacts on critical infrastructure, particularly electrical transmission grids. One of the foundational concepts of a grid's resilience is its robustness to extreme weather events, such as hurricanes. Resilience of the electric grid to high wind speeds is predicated upon the location and physical characteristics of the system components. Previous modeling assessments of electric grid failure were done at the systems level with assumptions on location and type of specific components. To facilitate more explicit adaptation metrics, accurate component‐level information is needed. In this study, we build and utilize a data set of location, physical characteristics, and age of transmission structures for nine counties in the Florida Panhandle. These component characteristics were then simulated for failure under a variety of scenarios using fragility curves. Eight hurricanes were modeled using Hazus from the Federal Emergency Management Administration and the resulting impact to the network was assessed. The network was generated using the transmission lines and towers, showing increasing impacts to network efficiency with larger storms. Although modern transmission structures are built under the more stringent extreme wind loading construction standards, the prevalence of older, wooden transmission structures throughout the region poses a substantial risk to reliable electricity transmission during tropical cyclone events from the Gulf of Mexico.

Performance and Security Considerations for Differential Privacy Model with Adaptive Metrics

Over time, the differential privacy model has undergone numerous extensions and implementations by various researchers. Recently, a notable advancement introduced two adaptive metrics aimed at achieving improved calibration of security and accuracy by considering the risk associated with statistical queries. The efficacy of this extension highlights its potential superiority over the original version under specific conditions. Building upon this progress, our paper presents further enhancements to the extended differential privacy model, focusing on selected aspects of its implementation. These modifications hold promise for enhancing the overall performance and security of the solution at a system level.

Multi-agent Reinforcement Learning-based Adaptive Heterogeneous DAG Scheduling

Static scheduling of computational workflow represented by a directed acyclic graph (DAG) is an important problem in many areas of computer science. The main idea and novelty of the proposed algorithm is an adaptive heuristic or graph metric that uses a different heuristic rule at each scheduling step depending on local workflow. It is also important to note that multi-agent reinforcement learning is used to determine scheduling policy based on adaptive metrics. To prove the efficiency of the approach, a comparison with the state-of-the-art DAG scheduling algorithms is provided: DONF, CPOP, HCPT, HPS, and PETS. Based on the simulation results, the proposed algorithm shows an improvement of up to 30% on specific graph topologies and an average performance gain of 5.32%, compared to the best scheduling algorithm, DONF (suitable for large-scale scheduling), on a large number of random DAGs. Another important result is that using the proposed algorithm it was possible to cover 30.01% of the proximity interval from the best scheduling algorithm to the global optimal solution. This indicates that the idea of an adaptive metric for DAG scheduling is important and requires further research and development.

Research on the Relationship between Industrial Restructuring and Economic Growth from a Low-Carbon Perspective: Overview and Prospects

In this study, a comprehensive examination of pertinent literature concerning the determinants and metrics of industrial structural adaptation has enabled us to distill the intricate interplay between industrial structural adjustment and economic growth. At present, the primary findings have coalesced around a deeper understanding of the intricate interplay between these two factors, shedding light on the pivotal direction and determinants of industrial structural adaptation amidst the ever-evolving landscape of economic development. The present study focuses on the relationship between the low-carbon perspective and identifies that low-carbon factors primarily influence the relationship between industrial restructuring and economic growth in five key aspects. This paper focuses on exploring the relationship between industrial restructuring and economic growth under a low-carbon perspective, revealing that low-carbon factors predominantly influence their interaction in five key dimensions. Lastly, this study provides a forward-looking analysis on how to effectively conduct industrial restructuring for achieving sustainable low-carbon economic development in the future, aiming to stimulate further research in related fields.

A new heuristic algorithm based on multi-criteria resilience assessment of human–robot​ collaboration disassembly for supporting spent lithium-ion battery recycling

Recycling spent lithium-ion batteries is a significant way to achieve life-cycle management and a green circular economy, helping to achieve carbon neutrality. The form of battery pack disassembly has gradually moved away from manual manipulation to a human–robot collaborative process. However, there is no established methodology for assessing spent lithium-ion battery disassembly between the environmental benefits, technical feasibility, and economic viability of human–robot collaborative disassembly (HRCD). To ensure the efficiency and security of HRCD processes, this paper develops a resilience assessment model for HRCD using stability, redundancy, efficiency and adaptation metrics. The resilience assessment theory is used to obtain the internal relations among the resilience indicators of the system. In this work, we evaluate the resilience of HRCD for lithium-ion battery disassembly by integrating fuzzy Bayesian fusion with an analytical network process unfolding cloud. The 2017 Chevrolet Bolt driver demonstrated the feasibility and effectiveness of the proposed approach, which allows greater flexibility in complex tasks performed by humans and robots. In case study, the weights values of stability, redundancy, efficiency and adaptability are 0.340533, 0.161384, 0.203835 and 0.294248, respectively. IR2&HO1 and IR3&HO2 have an assessment value of 93 and 95, respectively, with eigenvalues of 1.28 and 1.23, which are level I. RI3&HO1 and RI4&HO3 belong to level IV, with an evaluation value of 38 and 45, respectively, and a characteristic value of 2.27 and 2.55, respectively. The integrated methodology proposed in this paper could be effectively used by disassembly management, to evaluate the HRCD disassembly scheme resilient strategies for better managing production.

Adaptive metrics for an evolving pandemic: A dynamic approach to area-level COVID-19 risk designations

Throughout the COVID-19 pandemic, policymakers have proposed risk metrics, such as the CDC Community Levels, to guide local and state decision-making. However, risk metrics have not reliably predicted key outcomes and have often lacked transparency in terms of prioritization of false-positive versus false-negative signals. They have also struggled to maintain relevance over time due to slow and infrequent updates addressing new variants and shifts in vaccine- and infection-induced immunity. We make two contributions to address these weaknesses. We first present a framework to evaluate predictive accuracy based on policy targets related to severe disease and mortality, allowing for explicit preferences toward false-negative versus false-positive signals. This approach allows policymakers to optimize metrics for specific preferences and interventions. Second, we propose a method to update risk thresholds in real time. We show that this adaptive approach to designating areas as "high risk" improves performance over static metrics in predicting 3-wk-ahead mortality and intensive care usage at both state and county levels. We also demonstrate that with our approach, using only new hospital admissions to predict 3-wk-ahead mortality and intensive care usage has performed consistently as well as metrics that also include cases and inpatient bed usage. Our results highlight that a key challenge for COVID-19 risk prediction is the changing relationship between indicators and outcomes of policy interest. Adaptive metrics therefore have a unique advantage in a rapidly evolving pandemic context.

Graph Representation Learning With Adaptive Metric

Contrastive learning has been widely used in graph representation learning, which extracts node or graph representations by contrasting positive and negative node pairs. It requires node representations (embeddings) to reflect their correlations in topology, increasing the similarities between an anchor node and its positive nodes, or reducing the similarities with its negative nodes in embedding space. However, most existing contrastive models measure similarities through a fixed metric that equally scores all sample pairs in a specific feature space, but ignores the varieties of node attributes and network topologies. Moreover, these fixed metrics are always defined explicitly and manually, which makes them unsuitable for applying to all graphs and networks. To solve these problems, we propose a novel graph representation learning model with an adaptive metric, called GRAM, which produces appropriate similarity scores of node pairs according to the different significance of each dimension in their embedding vectors and adaptive metrics based on data distribution. With these scores, it is better to train a graph encoder and obtain representative embeddings. Experimental results show that GRAM has strong competitiveness in multiple tasks.

Functional adaptability in playing style: A key determinant of competitive football performance

Purpose The present study examined the relationship between playing style adaptability and team match performance indicators throughout the season. Three playing style adaptability metrics were analysed, namely, (1) flexibility (i.e., exhibiting a wide range of playing styles), (2) reactivity (i.e., adapting playing style based on opposition) and (3) imposition (i.e., executing predetermined playing style regardless of opposition). Methods Team playing styles were derived through a clustering analysis of 21,708 matches played in the top five male European leagues from 2014/15 to 2019/20. Spearman’s correlation was utilized to assess the association between the three playing style adaptability metrics and four team match performance indicators (e.g., shots taken in opposition penalty box; shots conceded in own penalty box; goals scored; goals conceded; and total wins). Results Playing style flexibility was positively associated with both offensive and defensive match performance indicators and win frequency. Conversely, playing style reactivity and imposition were negatively associated with these team match performance indicators. Conclusions Our results suggest that the capacity to exhibit a wide range of playing styles throughout a season is associated with greater team performance. Furthermore, it is possible that high performing teams are capable of functionally switching between playing style reactivity and imposition, depending on match dynamics.

Data optimization for the knowledge bases in the oil and gas Monitoring-While-Drilling (MWD) Systems

The ability of a knowledge-oriented system to learn requires the selection and organization of data. The article is devoted to the disclosure of the stages of data preparation for the application of case-based procedures to support decision-making by the operator of the technological process in the field of oil and gas wells drilling. The presented concept allows concentrating expert experience in the subject area and satisfies the criteria of the main sources of knowledge required for the functioning of case-based considerations. The main potential of artificial intelligence to learn is based on semantic dictionaries of the subject area, case databases in the form of knowledge bases, metrics of similarity, adaptation and configuration containers, which are identified as the sources of knowledge. The effectiveness of using computer simulators as a platform for experimental research of new models of complex technological processes is estimated.

Investigation of deep transfer learning for cross-turbine diagnosis of wind turbine faults

Data-driven fault diagnosis of wind turbines has gained popularity, and various deep learning models have been developed accordingly with massive amounts of data and achieved an excellent diagnosis performance. However, most existing deep learning models require a similar distribution of both training and testing data, thus the trained model cannot generalize new wind turbines with different data distributions. In addition, there are insufficient fault data in practice, and therefore the cost of training a new model from scratch is extremely high. To solve these problems, a cross-turbine fault diagnosis method based on deep transfer learning is proposed for wind turbines with the available supervisory control and data acquisition (SCADA) data. To better capture the spatial features of SCADA data, a deep multi-scale residual attention convolutional neural network (DMRACNN) is first designed. Then, the distribution differences between the source and target domain data are aligned at feature level. Specifically, we investigate the transfer performance of four different domain adaptation metrics. We evaluate our proposed method using SCADA data from two wind turbines to compare the diagnostic performance of four basic networks combined with four transfer metrics. Compared with traditional deep learning methods, our proposed DMRACNN achieved significant performance improvements. A cross-validation experiment using two turbines demonstrates the strong generalization ability of the proposed method.