Diverse Constraints Research Articles

Consensus and diversity-fusion partial-view-shared multi-view learning

Due to the multi-perspective of data, multi-view learning (MVL) is usually employed. Although existing MVL approaches gain fruitful achievements, they may neglect to learn (a) partial-view-shared knowledge between views; and (b) differences across views. Consequently, inadequate complementary knowledge and weak discriminability may be gained. To address the above problems, Consensus and Diversity-fusion Partial-view-shared Multi-view Learning (CDPMVL) is proposed, which includes two components: (a) Consensus, Partial-view-shared and Specific Component Learning (CPSCL) that partitions the samples into consensual, partial-view-shared, and specific parts, and learns the consensual, partial-view-shared, and specific knowledge of views; and (b) Diversity-fusion Partial-view-shared Knowledge Enhancement (DPKE) that imposes a diversity constraint on partial-view-shared parts and employs a heuristic-based auto-weighting mechanism to highlight the differences among views. By CDPMVL, more complementary relationships between and across views are explored, and the discriminability of the model is enhanced. Extensive experiments performed with eleven algorithms on nine datasets verify the superiority of CDPMVL, which indicates that the incorporation of partial-view-shared knowledge indeed enhances the complementary ability of views. The source code of CDPMVL is available at https://github.com/zzf495/CDPMVL.

Considering representation diversity and prediction consistency for domain generalization semantic segmentation

Existing domain generalization semantic segmentation (DGSS) methods aim to learn domain-invariant representation from single or multiple domains by using the consistency constraint or normalization strategy. Although these methods can improve the generalization capability of the models, the representation capability may be inevitably weakened since some effective information is eliminated, which may lead to the low discrimination capability of models. To simultaneously preserve the generalization and discrimination capability of models, we provide a novel perspective that encourages to learn the distinct representation while keeping the consistent prediction. Based on this, we propose a novel approach for domain generalization semantic segmentation by simultaneously considering the representation diversity and prediction consistency. The proposed approach consists of an adaptive style randomization (ASR) module, a representation diversity (RD) constraint, and a prediction consistency (PC) constraint. Specifically, first, since diverse stylized images are beneficial for the RD constraint to learn the distinct representation, the ASR module is proposed to leverage global and local style randomization to generate diverse stylized images which have the same content as the source domain images. Then, the distinct representation is learned from the source domain images and the stylized images by using the RD constraint to improve the discrimination capability of the model. Finally, the distinct representation is expected to generate consistent prediction by using the PC constraint for preserving the generalization capability of the model. Extensive experiments demonstrate that our approach achieves superior performance over current approaches on several benchmarks.

A novel solution for routing a swarm of drones operated on a mobile host

The increasing use of drones across various sectors demands optimized deployment strategies under diverse constraints. This paper tackles the Multiple Capacitated Mobile Depot Vehicles Routing Problem (mCMoD-VRP), a challenging variant of the Vehicle Routing Problem (VRP) where multiple drones with limited flight range operate from a mobile depot. The goal is to maximize target coverage while considering flight endurance, depot mobility, and drone multiplicity. We introduce a novel evolutionary algorithm, Evolutionary Optimization for Synchronized Routing Problem (EOSRP), which constructs synchronized routes for the drone swarm, accounting for all constraints. EOSRP distinguishes itself with specialized genetic operators, specifically designed to efficiently handle the constraints of mCMoD-VRP, enhancing both exploration and exploitation of the search space. EOSRP also facilitates collaborative planning among drones, enabling them to share targets and optimize routes collectively, resulting in more efficient use of flight range capacity. Comprehensive simulations on benchmark problems demonstrate that EOSRP consistently outperforms a serialized version of our previous single-drone algorithm, Genetic Algorithm for Capacitated Mobile Depot (GA-CMoD), achieving an average of 8.7% higher target coverage and 7.28% more efficient use of flight range capacity. EOSRP’s ability to generate synchronized solutions through collaborative planning leads to significantly improved mission efficiency.

General output constraints control of interconnected nonlinear systems with sensor faults

This paper delves into the problem of generalized output-constrained control for interconnected nonlinear systems, where sensor faults are modeled as time-varying unknown functions. First, a barrier function related to an output transformation function is proposed, which can meet the diverse constraint requirements by adjusting some parameters and functions. Second, the sensor fault is described as an unknown time-varying function, and its difficulties to the controller are successfully overcome with the help of adaptive control technology. Then, an output related function is introduced into the controller to handle the interconnected functions without approximators, while retaining the uncertainties in the interconnected functions. Based on the dynamic surface control technology, an adaptive general output constraint fault-tolerant controller is proposed. With the aid of the Lyapunov stability theory, it rigorously proves that the output satisfies the pre-given constraint requirements, and all signals of the closed-loop systems are bounded. Finally, the numerical simulation is presented to illustrate the validity of the proposed algorithm.

Perceived effects of climate change on aquaculture production in Zambia: status, vulnerability factors, and adaptation strategies

Aquaculture plays a critical role in Zambia's food and nutrition security. However, aquaculture is highly vulnerable to the effects of a changing climate which can lead to economic losses, and food and nutrition insecurity. In this study, we investigated the perceived effects of climate change on aquaculture production in Zambia, vulnerability factors, and adaptation strategies. We used semi-structured and structured questionnaires to collect data from aquaculture producers in all provinces across the country. Results revealed high (over 80%) awareness of climate change, primarily attributed to literacy levels and access to media. Producers perceived occurrence of climate-related changes such as rising temperature, altered rainfall patterns, cyclones frequency, and disease outbreaks, impacting production costs. Generally, the results indicate a high to medium perception of rising temperature and increased frequency of flooding and droughts across the country. Although rising temperature did not appear to affect aquaculture production cost, the frequency of droughts, floods, cyclones, and disease outbreaks showed a significant association with production costs. Identified vulnerability factors include reliance on a limited range of fish species, absence of insurance coverage, and low adoption of adaptation practices. Vulnerability levels vary across regions, with provinces in agroecological zone I (Eastern, Lusaka, Southern, and Western) showing higher vulnerability. Despite this, many producers aren't implementing adaptation measures due to financial limitations, species diversity constraints, and insufficient knowledge of alternative economic pursuits. Main adaptation strategies involve livelihood diversification and adjustments in fish cultivation periods and infrastructure. To foster aquaculture sustainability amid climate change, critical interventions such as farm insurance, research diversification in aquaculture species, and enhancing producer resilience are necessary.

Construction and validation of computational techniques as part of a holistic approach to complex engineering design

As a full-service provider of bespoke solutions to environmental noise problems, there is a need to have a precise understanding of the performance of our engineered products and complex assemblies. More specifically, there are risks, as well as limitations, to relying solely on a single scientific method-e.g., empirical, computational, numerical or analytical. In the following paper, we describe the validation process of sophisticated computational software-a suite of Siemens software-to bolster our risk assessment practices. Although computational tools generally require extensive 'calibration,' our interest in pursuing the software is to model diverse constraints and environmental conditions of projects, which cannot be 'simply' assessed using empirical, numerical and analytical tools. Herein are presented examples of projects utilizing our holistic approach to assuring the performance of our engineering solutions. More specifically, we employ aero-vibro-acoustical simulations to model the aero-vibro-acoustical performance (e.g., airflow and pressure drop, dynamic insertion loss) of our industrial acoustic silencers (and similar complex assemblies).

You only compress once: Towards effective and elastic BERT compression via exploit–explore stochastic nature gradient

Despite superior performance on various natural language processing tasks, pre-trained models such as BERT are challenged by deploying on resource-constraint devices. Most existing model compression approaches require re-compression or fine-tuning across diverse constraints to accommodate various hardware deployments. This practically limits the further application of model compression. Moreover, the ineffective training and searching process of existing elastic compression paradigms (Wang et al., 2020; Cai et al., 2020) prevents the direct migration to BERT compression. Motivated by the necessity of efficient inference across various constraints on BERT, we propose a novel approach, YOCO-BERT, to achieve compress once and deploy everywhere. Specifically, we first construct a huge search space with 1013 architectures, which covers nearly all configurations in BERT model. Then, we propose a novel stochastic nature gradient optimization method to guide the generation of optimal candidate architecture which could keep a balanced trade-off between explorations and exploitation. When a certain resource constraint is given, a lightweight distribution optimization approach is utilized to obtain the optimal network for target deployment without fine-tuning. Compared with state-of-the-art algorithms, YOCO-BERT provides more compact models, yet achieving 2.1%–4.5% average accuracy improvement on the GLUE benchmark. Besides, YOCO-BERT is also more effective, e.g., the training complexity is O(1) for N different devices. Codes available https://github.com/MAC-AutoML/YOCO-BERT.

Optimizing the performance of a new current source type grid-forming modular multilevel converter using genetic algorithms

Modular Multilevel Converters (MMC) have found extensive application in connecting Renewable Energy Sources to the power grid. Recently, MMCs operated in the Grid Forming (GFM) mode have garnered significant interest due to their superior performance in weak AC networks. The Virtual Synchronous Generator (VSG) is a GFM technology designed to emulate the behavior of a real synchronous generator (SG). There is flexibility in selecting the VSG’s parameters as they are not constrained by the physical attributes of an actual SG. This paper proposes employing the Genetic Algorithm (GA) to select these parameters, underscoring its simplicity and effectiveness in meeting diverse constraints and objectives. Small Signal (SS) models are constructed at potential operational points, with GA leveraging the SS model’s eigenvalues to select parameter values that enhance dynamic performance, ensuring resilience across a wide range of operating conditions. The effectiveness of these designed parameters is validated through Electromagnetic Transients (EMT) simulations.

Source-free cross-domain fault diagnosis of rotating machinery using the Siamese framework

Despite deep learning based intelligent diagnosis has become an essential means of perceiving the health status of rotating machinery, existing diagnostic models struggle to efficiently extract domain-invariant representations for cross-domain tasks in the absence of source domain data due to the limitations of imbalanced machine fault data, the various scenarios, as well as privacy protection in industrial applications. In order to solve these issues, we innovatively propose a source-free cross-domain fault diagnosis algorithm for rotating machinery using the Siamese framework (SCSF). Specifically, a deep residual shrinkage network with parallel spatial and channel-wise attention mechanisms (DRSN-SCW) is used for feature extraction to suppress background noise. In addition, pre-trained feature extraction networks and classifiers use only source domains based on the Siamese framework for end-to-end representation learning to address the problem of fault diagnosis of category-imbalanced data at the feature representation level. Finally, for cross-domain scenarios where the target domain does not have access to source domain data during training, we use a prototype-based pseudo-labeling strategy as well as impose consistency and diversity constraints on the classifier outputs to improve the diagnostic performance of unlabeled target domains. Extensive experiments conducted on the Spectra Quest mechanical fault dataset and the Case Western Reserve University rolling bearing dataset validate the effectiveness of the SCSF. The code is available at .

‘Finding the play’ – exploring with occupational therapists practice possibilities in the context of Irish schoolyards

Background Research has identified diverse constraints to the adoption of school-based occupational therapy approaches and a lack of attention to addressing the barriers to children’s play opportunities. Critical contextualised research is advocated to inform practice possibilities. Aims/Objectives This inquiry aimed to explore with occupational therapists their existing practices in Irish schoolyards to generate practice possibilities concerned with play, as an issue of occupational justice. Materials and Methods Using the theory of practice architectures, six occupational therapists from diverse sites of practice participated in the first phase of a critical action research process using dialogical focus group and occupational mapping methods. Results Three themes were generated (1) Existing practices as situated (2) (Re)mattering play and practices as occupations and (3) Practice possibilities – ‘Finding the play’ between responsiveness and responsibilities. A further interrelated dimension was how the research methods provided mechanisms of raising consciousness. Conclusions, and Significance Alongside constructing knowledges on existing practices in an Irish context, this inquiry contributes to understandings of practices as socially embedded generative processes of ‘finding the play’, highlighting ethical responsibilities to make visible inequities reproduced in habitual practices and engage in relationships of solidarity to (re)construct alternative shared practices.

A Novel Constraint-Aware Flexible Model with Ant Colony Optimization for Symmetrical Travel Recommendation

This paper proposes a flexible travel recommender model (FTRM) that emphasizes the symmetry between user preferences and travel constraints, addressing key challenges in the field such as the integration of diverse constraint types and the customization of travel itineraries. The key contribution of the proposed model lies in its integration with the item constraints data model (ICDM), which effectively manages a plethora of constraint types. Additionally, this study develops a novel algorithm inspired by ant colony optimization (ACO) principles, demonstrating performance metrics that are comparable to state-of-the-art algorithms in this field. A comprehensive set of systematic experimental analyses is conducted, employing various models across diverse situational contexts, with the primary goal of illustrating the capabilities of the proposed symmetrical FTRM using real-world data from the Durham dataset. The obtained results highlight the model’s ability to accommodate diverse constraint types, facilitating the customization of travel itineraries to suit individual user preferences and achieve a balanced and symmetrical travel experience. Specifically, our model outperforms existing models in terms of flexibility and customization, showing significant improvements in user satisfaction and itinerary efficiency.

Inverse Adversarial Diversity Learning for Network Ensemble.

Network ensemble aims to obtain better results by aggregating the predictions of multiple weak networks, in which how to keep the diversity of different networks plays a critical role in the training process. Many existing approaches keep this kind of diversity either by simply using different network initializations or data partitions, which often requires repeated attempts to pursue a relatively high performance. In this article, we propose a novel inverse adversarial diversity learning (IADL) method to learn a simple yet effective ensemble regime, which can be easily implemented in the following two steps. First, we take each weak network as a generator and design a discriminator to judge the difference between the features extracted by different weak networks. Second, we present an inverse adversarial diversity constraint to push the discriminator to cheat generators that all the resulting features of the same image are too similar to distinguish each other. As a result, diverse features will be extracted by these weak networks through a min-max optimization. What is more, our method can be applied to a variety of tasks, such as image classification and image retrieval, by applying a multitask learning objective function to train all these weak networks in an end-to-end manner. We conduct extensive experiments on the CIFAR-10, CIFAR-100, CUB200-2011, and CARS196 datasets, in which the results show that our method significantly outperforms most of the state-of-the-art approaches.

Optimizing Cloud Computing Resources: An Energy Efficient Multi-QoS Factor-Based VM Placement Strategy

Cloud computing is experiencing unprecedented demand, offering scalable and flexible resources for a wide range of applications. However, this surge in demand has raised concerns about energy consumption and the need for environmentally sustainable solutions. Green computing has emerged as a critical consideration in this context. Virtual Machine Placement (VMP) is a key component of optimizing cloud resources, aiming to allocate virtual machines efficiently while minimizing energy consumption, cost, and load balancing. This paper addresses the VMP problem by introducing a novel approach based on multifactor optimization, specifically the Diversity Constraint Digger Snake Optimizer (DCD-SO). It offers an innovative perspective on optimizing virtual machine placement by considering energy efficiency, load balancing, and resource utilization simultaneously with the aim to reduce VM migration count, time and cost. Proposed method provides a more comprehensive and sustainable solution, aligning with the principles of green computing. Through extensive simulations and experiments, we have rigorously evaluated the performance of DCD-SO in comparison to traditional optimization techniques such as Particle Swarm Optimization (PSO) and Snake Optimization. In our analysis of actual cloud environments, we compared the results of our method with existing state-of-the-art techniques. Result outcomes determine showed that proposed approach has reduced migration count of 5 and 3 for scheduling 42VMs and 84VMs on 16 and 32 host units respectively than traditional MOGANS, GA-S, GA-N and GA-NN methods. This comprehensive evaluation reinforces the effectiveness and practicality of our approach in addressing the intricate challenges of Virtual Machine Placement (VMP) in dynamic cloud computing settings. As cloud computing continues to evolve, our study contributes to more sustainable and efficient resource management, addressing both current demands and future needs.

Intelligent mission planning of sky survey based on the NSGA-III algorithm

In recent years, China has been scheduled to launch an all-sky survey telescope for celestial observation. The meticulous arrangement of observation sequences for this telescope is of paramount importance, given the myriad challenges and constraints inherent to such astronomical survey missions. These challenges encompass factors such as spatial limitations, stray light contamination, as well as constraints related to data storage and energy consumption. This undertaking presents a quintessential NP-hard optimization problem. To address this complexity and enhance in-orbit observation efficiency, this study closely adheres to practical engineering considerations. A comprehensive evaluation of diverse constraints is conducted, culminating in the establishment of an optimization model for systematic planning of sky survey missions. The primary optimization objectives encompass maximizing in-orbit observation duration, augmenting the count of observable celestial points, and minimizing maneuvering angles. To tackle this multifaceted problem, a mission planning algorithm is devised based on NSGA-III, a multi-objective optimization technique. The experimental results indicate that the sky survey planning algorithm proposed in this paper can effectively accomplish sky survey observation tasks, achieving the scientific objective of no fewer than 350 daily observation points. It holds the potential to provide algorithmic support for the sky survey telescope mission on the Chinese space station.

A pareto fronts relationship identification-based two-stage constrained evolutionary algorithm

Striking a balance between diverse constraints and conflicting objectives is one of the most crucial issues in solving constrained multi-objective optimization problems (CMOPs). However, it remains challenging to existing methods, due to the reduced search space caused by the constraints. For this issue, this paper proposes a Pareto fronts relationship identification-based two-stage constrained evolutionary algorithm called RITEA, which balances objective optimization and constraint satisfaction by identifying and utilizing the relationship between the unconstrained Pareto front (UPF) and the constrained Pareto front (CPF). Specifically, the evolutionary process is divided into two collaborative stages: training stage and reinforcement stage. In the training stage, a relationship identification method is developed to estimate the relationship between UPF and CPF, which guides the population search direction. In the reinforcement stage, the corresponding evolutionary strategies are designed based on the identified relationship to enhance the accurate search on the CPF. Furthermore, a dynamic preference fitness function (termed DPF) is designed to adaptively maintain the balance of search preference between convergence and diversity. Compared to seven state-of-the-art algorithms on 36 benchmark CMOPs in three popular test suites, RITEA obtains 77.8% of the best IGD values and 66.7% of the best HV values. The experimental results show that RITEA exhibits highly competitively when dealing with CMOPs.

Exploring with children, play in Irish primary schoolyards

ABSTRACT Children’s play in Irish schoolyards remains neglected in educational policies and practices despite government commitments to inclusive schools and children’s rights. There is a dearth of research on children’s perspectives of play, criticisms of ‘at risk’ discourses underpinning concerns for certain children’s play rights, and studies identifying exclusion within Irish schoolyards, particularly for children with minoritized identities. This inquiry informed by the theory of practice architectures used walking interviews to explore with twenty-three children their play practices in two Irish primary schools identified as disadvantaged. Analysis of the interviews generated three themes: (1) the state of play – cracks with(in) the routines of the schoolyard, (2) playing along and with(in) this shared space and (3) the hard yard. This inquiry contributes to understandings of children’s play with(in) Irish schoolyards, as socially situated practices with contrasting representations of play as habitual and emerging. Play was central to children’s social lives, identities, and friendships and interrelated with diverse constraints, exclusionary practices, and the (re)production of the ‘hard yard’. While mattering most children’s experiences of significant constraints and inequities, this inquiry also highlighted the transformative possibilities generated within play to create shared possibilities for individual and collective flourishing.

Analysis of Reliability and Cost of Complex Systems with Metaheuristic Algorithms

Introduction. The assessment of the reliability and cost of complex systems, such as Complex Bridge Systems (CBS) and Life Support Systems in Space Capsules (LSSSC), is fascinating. To achieve the ideal system design through diverse constraints and increase overall system reliability, researchers have extensively explored system reliability and cost optimization problems. Hence, the significant advancement in metaheuristic methods is the primary source of further system reliability and cost optimization process refinement.
 Aim and tasks. This research attempts to enhance the reliability and cost of complex systems named CBC and LSSSC has been presented. 
 Results. The structure is based on few recent metaheuristic techniques, such as Moth Flame Optimization (MFO), Whale Optimization Algorithm (WOA), Gazelle Optimization Algorithm (GOA), Dragonfly Algorithm (DA), and Coati Optimization Algorithm (COA). Comparing the acquired findings to those found in other proposed techniques demonstrates the usefulness of a methodology based on COA. The proposed COA algorithm exhibits enhanced efficiency by offering superior solutions to reliability and cost-optimization problems. In addition, a non-parametric Friedman ranking was performed for validation. The results of this research are based on improving the reliability of the parameters and decreasing complex systems’ costs used by the five metaheuristic methods. Observing the convergence graph, Friedman ranking, statistical results test, and tables determined that COA is the most effective algorithm for a complex system’s cost and reliability parameters compared to other existing approaches, and also provided a faster solution.
 Conclusions. This study proposes unique ways to reduce costs while increasing parameter reliability in complex systems. After analysing the comparative solution, the authors found that when comparing these approaches (GOA, DA, MFO, WOA, and COA), the COA provided the best minimum solution for the cost and reliability of complex systems. Hence, the suggested COA procedure was more successful than that described in this study.

Constraints to agricultural finance in underdeveloped and developing countries: a systematic literature review

ABSTRACT Agricultural finance plays a pivotal role in facilitating the modernization and commercialization of farming practices and bolstering global food security. However, the provision of sustainable financial services in remote regions poses a significant challenge within the developing and underdeveloped countries. Consequently, farmers often grapple with inadequate access to formal credit sources and essential productive assets due to diverse constraints. Although, several studies have investigated constraints to agriculture finance; the literature is divergent and a holistic framework of constraints to agriculture finance is missing. Therefore, this research explores the constraints to agriculture finance in 31 developing and underdeveloped countries using systematic review approach, and synthesizes a holistic framework of the constraints using Qualitative Evidence Synthesis (QES) method that can be used as dependable framework for conceptualization and policy formation. Moreover, it describes the interplay and mutual reinforcement between the challenges faced by different stakeholders to elaborate how they influence each other in complex ways. Additionally, this study suggests a collective and multi-faceted approach for overcoming barriers to agricultural financing, and highlights the limitations and unintended consequences of agricultural credit and its implications for agricultural sustainability.

Diversity-Authenticity Co-constrained Stylization for Federated Domain Generalization in Person Re-identification

This paper tackles the problem of federated domain generalization in person re-identification (FedDG re-ID), aiming to learn a model generalizable to unseen domains with decentralized source domains. Previous methods mainly focus on preventing local overfitting. However, the direction of diversifying local data through stylization for model training is largely overlooked. This direction is popular in domain generalization but will encounter two issues under federated scenario: (1) Most stylization methods require the centralization of multiple domains to generate novel styles but this is not applicable under decentralized constraint. (2) The authenticity of generated data cannot be ensured especially given limited local data, which may impair the model optimization. To solve these two problems, we propose the Diversity-Authenticity Co-constrained Stylization (DACS), which can generate diverse and authentic data for learning robust local model. Specifically, we deploy a style transformation model on each domain to generate novel data with two constraints: (1) A diversity constraint is designed to increase data diversity, which enlarges the Wasserstein distance between the original and transformed data; (2) An authenticity constraint is proposed to ensure data authenticity, which enforces the transformed data to be easily/hardly recognized by the local-side global/local model. Extensive experiments demonstrate the effectiveness of the proposed DACS and show that DACS achieves state-of-the-art performance for FedDG re-ID.

Contributing Dimension Structure of Deep Feature for Coreset Selection

Coreset selection seeks to choose a subset of crucial training samples for efficient learning. It has gained traction in deep learning, particularly with the surge in training dataset sizes. Sample selection hinges on two main aspects: a sample's representation in enhancing performance and the role of sample diversity in averting overfitting. Existing methods typically measure both the representation and diversity of data based on similarity metrics, such as L2-norm. They have capably tackled representation via distribution matching guided by the similarities of features, gradients, or other information between data. However, the results of effectively diverse sample selection are mired in sub-optimality. This is because the similarity metrics usually simply aggregate dimension similarities without acknowledging disparities among the dimensions that significantly contribute to the final similarity. As a result, they fall short of adequately capturing diversity. To address this, we propose a feature-based diversity constraint, compelling the chosen subset to exhibit maximum diversity. Our key lies in the introduction of a novel Contributing Dimension Structure (CDS) metric. Different from similarity metrics that measure the overall similarity of high-dimensional features, our CDS metric considers not only the reduction of redundancy in feature dimensions, but also the difference between dimensions that contribute significantly to the final similarity. We reveal that existing methods tend to favor samples with similar CDS, leading to a reduced variety of CDS types within the coreset and subsequently hindering model performance. In response, we enhance the performance of five classical selection methods by integrating the CDS constraint. Our experiments on three datasets demonstrate the general effectiveness of the proposed method in boosting existing methods.