Large-scale multimodal multiobjective optimization problems with sparse Pareto optimal solutions pose a significant challenge in the field of optimization, primarily stemming from the multimodality characteristic, the curse of dimensionality, and the uncertainty of sparse solutions. In this work, we propose a sparse large-scale multimodal multiobjective evolutionary algorithm (MVDE-MMEA) to solve such complex tasks. In MVDE-MMEA, a multiview diversity enhancement mechanism is designed to improve the exploration ability of the algorithm across the entire decision space. The diversity mechanism seamlessly transits from a global perspective to a local one as the population evolves, which contributes to an effective convergence towards multiple Pareto optimal sets in large-scale decision space. Furthermore, a clustering method is proposed to divide the population into several niches, guided by the shared characteristics among candidates. In this way, the algorithm shifts its focus from exploration in the whole search space to exploitation in local regions. Experimental studies are conducted on eight benchmark test problems and 12 feature selection problems. The comparative results with the state-of-the-art algorithms demonstrate the superiority of MVDE-MMEA.

With the rapid development of industrial Internet of Things (IIoT), the edge-cloud collaboration architecture combining the powerful computing ability of cloud computing with the low latency of edge computing plays an increasingly important role in providing the computing resources and reducing the latency for IIoT. However, under this architecture, existing methods in scheduling the resources for IIoT often focus on latency and energy consumption but ignore some other important factors especially the reliable factor, thereby making it difficult for them to adapt to real-world IIoT scenarios. To this end, we propose a reliable resource scheduling method with knowledge transfer for edge-cloud collaboration-enabled IIoT. Specifically, we first model the resource scheduling as a many-objective optimization problem, considering these optimized objectives: latency, energy consumption, load balance, resource utilization, and trust measure between tasks and servers. Then, we develop a knowledge transfer accelerated clustering evolutionary algorithm (KTCEA) for many-objective optimization to solve the model, where the knowledge transfer aims at accelerating the evolution and the clustering makes the population converge from various directions. Under the collaboration of knowledge transfer and clustering, KTCEA can utilize the small population size to effectively search the objective space, and thus have the high real-time performance. Extensive experiment results on a benchmark test suite and the constructed model demonstrate that KTCEA is highly competitive compared with some advanced methods and our method can efficiently achieve the resource scheduling for edge-cloud collaboration-enabled IIoT, respectively.

Multimodal Emotion Recognition in Conversations (MERC) identifies utterance emotions by integrating both contextual and multimodal information from dialogue videos. Existing methods struggle to capture emotion shifts due to label replication and fail to preserve positive independent modality contributions during fusion. To address these issues, we propose a Dual Contrastive Learning Framework (DCLF) that enhances current MERC models without additional data. Specifically, to mitigate label replication effects, we construct context-aware contrastive pairs. Additionally, we assign pseudo-labels to distinguish modality-specific contributions. DCLF works alongside basic models to introduce semantic constraints at the utterance, context, and modality levels. Our experiments on two MERC benchmark datasets demonstrate performance gains of 4.67%-4.98% on IEMOCAP and 5.52%-5.89% on MELD, outperforming state-of-the-art approaches. Perturbation tests further validate DCLF’s ability to reduce label dependence. Additionally, DCLF incorporates emotion-sensitive independent modality features and multimodal fusion representations into final decisions, unlocking the potential contributions of individual modalities.

For sentiment analysis (SA) issue, most current SA models focus on Explicit Sentiment Analysis (ESA), with less attention to Implicit Sentiment Analysis (ISA). Recent ISA models cannot fully consider prior knowledge in Pre-trained Language Model (PLM) for this knowledge intensive task, even introducing complex external knowledge bases. This also leads to limited performance with additional computational resources. To address these problems, we propose a Prompt-based Implicit sentiment analysis Network (PIN) for Chinese ISA, where a topic recognition module is introduced to identify the topic of the review. Then, the topic is embedded in the soft template to predict sentiment based on prompt learning, which can effectively activate PLM knowledge with low computational resources. Experiments conducted on three public datasets demonstrate the effectiveness of our model, as compared with state-of-the-art methods. Meanwhile, we also provide our identified topic as a supplement to the above datasets, forming three new datasets.

Federated Learning (FL) enables collaborative training while protecting the privacy of participant data. However, typical centralized FL structures are vulnerable to malicious client attacks. To mitigate such vulnerabilities, blockchain technology has been used to develop a decentralized FL framework. Yet, this approach leads to significant transmission and computation overhead. In order to improve robustness as well as efficiency, we propose Multi-Population Federated Learning (MPFL), which optimizes blockchain-based federated learning by incorporating operators of multi-population genetic algorithm (MPGA), such as selection, crossover, mutation and migration. Moreover, we introduce a committee consensus mechanism tailored to MPFL, where honest clients are elected to govern each population, thereby facilitating intraspecific and interspecific competition. Our experiments demonstrate that MPFL outperforms various aggregation algorithms in achieving superior robustness, and it reduces the costs of transmission and computation compared to the blockchain-based FL framework.

Most of the current research on dynamic multiobjective optimization problems (DMOPs) assumes that environmental changes can be detectable. However, undetectable changes are frequently encountered in real-world applications, which pose a serious challenge for the existing methods. Because undetectable changes can lead to the failure of change detection techniques, thereby making it difficult to adapt to environmental changes for most algorithms. Therefore, to effectively deal with DMOPs with undetectable changes, this work proposes a dual mutation-based dynamic multiobjective evolutionary algorithm (DM-DMOEA). The proposed DM-DMOEA incorporates the following two main components. First, based on the exploration level of the population, an adaptive selection strategy is proposed, which enables the adaptive identification of individuals for mutation. Second, a dual mutation scheme is developed, utilizing both the polynomial mutation and the Gaussian mutation. These mutation operations are applied on the selected individuals to generate the mutated individuals, allowing for diverse exploration in the search space. After conducting the above two strategies, the population will evolve by the evolutionary criterion of multiobjective optimization. As a result, the algorithm can effectively adapt to undetectable changes in the environment. Comprehensive empirical studies are conducted on different benchmark functions and a real-world application to evaluate the performance of DM-DMOEA. Experimental results have demonstrated that DM-DMOEA is competitive in tracking the Pareto front over time when facing undetectable changes.

Constrained multimodal multiobjective optimization problems (CMMOPs) are challenging in the field of optimization, requiring to consider the balance between the constraints and objectives, the balance between exploration and exploitation in the decision space and the objective space, and the balance of diversity between the decision space and the objective space. In this work, we propose a multitask-assisted evolutionary algorithm (CMMO-MTA) to achieve these balances. In CMMO-MTA, a tri-task multitasking framework is proposed, which contains one main task and two assisting tasks. The main task aims to solve the original CMMOP, and two assisting tasks are designed to transfer desired knowledge to the main task to achieve the first two balances. Furthermore, a space balance-based selection mechanism is proposed to ensure a balanced representation of solutions in both the decision space and the objective space, thereby striking the third balance. Experimental studies are conducted on 31 test problems and a real-world application to compare the proposed algorithm with seven state-of-the-art algorithms. The results demonstrate the superiority of CMMO-MTA in solving CMMOPs.

With the rapid development of 5G, numerous data is generated in the blockchain-enabled industrial Internet of Things (IIoT). Although these peers in the blockchain system have the storage ability, they are far from meeting the storage requirements of the generated data. In addition, all data is stored in the blockchain network, which is unfriendly to applications that require real-time information. To address the above storage problems, this article proposes a block storage optimization method for blockchain-enabled IIoT, whose core idea is to conditionally select some blocks to store in the cloud. This method firstly models the selection conditions of blocks as a many-objective optimization problem, where the selection conditions include using probability, storage cost, space occupation, and transmission cost. Then, a cascading selection-based evolutionary algorithm (CSEA) for many-objective optimization is developed to solve the model and thereby obtain the optimal blocks stored in the cloud, where CSEA adopts the diversity-first principle. Finally, CSEA is first compared with seven state-of-the-art methods on two benchmark test suites for validating its ability to obtain reliable experimental results, and then is used to solve the proposed model. The corresponding results demonstrate that CSEA has high competitiveness, and our method can effectively address the storage problems above. In summary, this article provides a novel method for addressing the storage problem in the blockchain-enabled IIoT.

Medium-scale expensive multiobjective optimization problems (EMOPs) present a significant challenge to most existing surrogate-assisted evolutionary algorithms (SAEAs). Because the algorithms must balance convergence and diversity with a limited number of fitness evaluations (FEs), while managing the uncertainty in surrogate predictions within the medium-scale decision space. Therefore, this work proposes a surrogate-assisted multiobjective evolutionary algorithm based on two-level model management (SAMOEA-TL2M) to effectively address medium-scale EMOPs. In SAMOEA-TL2M, infill solutions are selected using the proposed two-level model management strategy. In the first level, the estimated non-dominated solutions with good shift-based density estimation (SDE) values are selected for the balance of convergence and diversity. In the second level, the estimated non-dominated solutions and high uncertainty solutions are considered. To quantify uncertainty, an inverse distance weighting (IDW) is introduced. Moreover, an accuracy rate indicator (ARI) is proposed for the optimization state assessment, providing guidance for adaptively executing the two model management levels. Extensive experiments on three widely used instances and time-varying ratio error estimation (TREE) problems with up to 120 dimensions demonstrate the superiority of SAMOEA-TL2M over five state-of-the-art SAEAs in solving medium-scale EMOPs.

Many problems related to natural language processing are solved by neural networks and big data. Researchers have previously focused on single-task supervised goals with limited data management to train slogan classification. A multi-task learning framework is used to learn jointly across several tasks related to generating multi-class slogan types. This study proposes a multi-task model named slogan generative adversarial network systems (Slo-GAN) to enhance coherence and diversity in slogan generation, utilizing generative adversarial networks and recurrent neural networks (RNN). Slo-GAN generates a new text slogan-type corpus, and the training generalization process is improved. We explored active learning (AL) and meta-learning (ML) for dataset labeling efficiency. AL reduced annotations by 10% compared to ML but still needed about 70% of the full dataset for baseline performance. The whole framework of Slo-GAN is supervised and trained together on all of these tasks. The text with the higher reporting score level is filtered by Slo-GAN, and a classification accuracy of 87.2% is achieved. We leveraged relevant datasets to perform a cross-domain experiment, reinforcing our assertions regarding both the distinctiveness of our dataset and the challenges of adapting bilingual dialects to one another.

The incipient fault detection of a complex industrial process is a challenging problem for traditional dynamic detection methods. Traditional dynamic detection methods usually decouple the correlations among the variables and dynamic correlations simultaneously, which makes the two types of correlations mixed and may lead to performance deterioration in long-sequence dynamic detection. Some incipient faults may not change the amplitudes of process variables but change the long-sequence dynamic features. Based on the$T^{2}$statistic and matrix multiplication transformation ($T^{2}$S-MMT), traditional dynamic detection methods can detect many faults effectively. However, the$T^{2}$S-MMT can not effectively detect some incipient faults due to the above two types of correlations mixed. In order to overcome the shortcomings of$T^{2}$S-MMT and improve the detection ability of some incipient faults, this paper proposes a fault detection method based on the dynamic k-nearest neighbor model and Dual Control Chart (DKNN-DCC), which can improve the incipient fault detection performance by using long-sequence dynamic detection. The proposed method is verified by the Tennessee Eastman (TE) process and the continuously stirred tank reactor (CSTR) process. The experimental results show the effectiveness of the proposed method in incipient fault detection compared with traditional dynamic detection methods.Note to Practitioners—This paper presents a novel incipient fault detection method, which directly mines the long-sequence dynamic abnormal information from the process variable and overcomes the problem of some abnormal information being submerged in the$T^{2}$statistic calculated based on the matrix multiplication transformation. The proposed method can detect incipient faults that are not easily detected by some traditional methods and can help operators find the abnormal and avoid more serious losses. The structure of the proposed method jumps out of the frameworks of traditional dynamic detection methods, which is feasible to apply to different stable industrial processes.

In some real-world applications, the optimization problems may involve multiple design stages. At each design stage, the objective is incrementally modified by incorporating more decision variables and optimized. In addition, the fitness evaluations (FEs) are often highly costly. Such optimization problems can be called expensive incremental optimization problems (EIOPs). Despite their importance, EIOPs have not attracted much attention over the past few years. Since the objectives of different design stages are different but related, reusing the search experience from the past design stages is beneficial to the evolutionary search of the current design stage. Therefore, a surrogate-assisted differential evolution with knowledge transfer (SADE-KT) is proposed in this work, which aims to fill the current gap in solving EIOPs. The major merit of the proposed SADE-KT is its ability to seamlessly integrate knowledge transfer and the surrogate-assisted evolutionary search. In SADE-KT, a surrogate based hybrid knowledge transfer strategy is first proposed. This strategy makes it possible to reuse the knowledge captured from the past design stages by leveraging different knowledge transfer techniques. As a result, the convergence for the current design stage can be speeded up. Then, a two-level surrogate-assisted evolutionary search is developed to search for the optimum. Comprehensive empirical studies have demonstrated that the proposed algorithm works efficiently on EIOPs.

The evolution of multi-constellation Global Navigation Satellite Systems (GNSS) has presented an opportunity to enhance user positioning accuracy. However, practical constraints, such as limited receiver channels and power consumption, necessitate judicious satellite selection. Geometric Dilution of Precision (GDOP) serves as a critical indicator for optimizing positioning performance, but determining the subset with the optimal GDOP value involves solving an impractical combinatorial optimization problem. A compromise solution seeks to balance computational complexity while sacrificing some optimality. Consequently, finding an optimal combination of satellites with a low computational burden yet quasi-optimal GDOP value remains a challenge. In response to this challenge, we introduce a pioneering approach in this paper: a super-fast satellite selection algorithm based on power series expansion (SF-PSE). This paper derives a “Xiao-Liang formula” based on power series expansion and the Sherman-Morrison formula. Using this formula, we propose two low-computational-cost rapid iterative algorithms (F-PSE and SF-PSE). Among these algorithms, SF-PSE notably reduces the computational burden through an approximate iterative inverse matrix-guided search method. Experimental results demonstrate that satellite combinations identified by F-PSE and SF-PSE yield nearly the same accuracy while reducing computation time by 70% and 87%, respectively, compared to the SMALLER method.

Surrogate-assisted evolutionary algorithms (SAEAs) are well suited for computationally expensive optimization. However, most existing SAEAs only focus on low- or medium-dimensional expensive optimization. Thus, a novel SAEA for high-dimensional expensive optimization, denoted as surrogate-assisted multipopulation particle swarm optimizer (SA-MPSO), is proposed and fully investigated in this work. The proposed algorithm employs a parameter-free clustering technique, denoted as affinity propagation clustering, to generate several subswarms. A surrogate-assisted learning strategy-based particle swarm optimizer is proposed for guiding the search of each subswarm. Furthermore, a model management strategy is adapted to choose the promising particles for real fitness evaluations. Finally, a subswarm diversity maintenance scheme and a surrogate-based trust region local search technique are introduced to enhance both exploration and exploitation. The experimental results on commonly used benchmark test problems with dimensions varying from 30 to 100 and airfoil design problem have shown that SA-MPSO outperforms some state-of-the-art methods.

Dominance robustness (DR) has been proposed for assessing the ability of the Pareto-optimal solutions to remain to be nondominated when the decision variables are subject to noise. There are two main challenges in search for dominance robust optimal solutions in dominance robust multiobjective optimization (MOP), namely, accurate estimation of the DR measure and a good balance between convergence and DR in the presence of uncertainty. In this article, a novel robust MOP evolutionary algorithm based on decision variable assortment (DVA) is proposed to tackle these challenges. To be specific, an indicator, termed as dominance robust indicator, is proposed to measure the DR based on the dominance level and dominance relationship of the sampled points. Then, the decision variables are divided into low DR-related variables and high DR-related variables based on the DVA strategy. Finally, low and high DR related variables are optimized separately to obtain the dominance robust optimal solutions. In addition, performance indicators to quantify the performance of dominance robust optimal solution set obtained by robust MOP algorithm are proposed. Experimental results have demonstrated that the proposed algorithm is competitive in search for dominance robust optimal solutions.

In recent years, surrogate-assisted evolutionary algorithms have been developed for expensive optimization. However, a majority of applications are dynamic optimization problems in the real-world. In this paper, therefore, a surrogate-assisted clustering particle swarm optimizer is proposed for expensive dynamic optimization. In the proposed method, several clusters are first created by affinity propagation clustering, and then local radial basis function (RBF) surrogates are built based on the neighbor evaluated points for each cluster. Finally, in each cluster, the local RBF assists particle swarm optimizer to search the most promising point, which is evaluated by real objective function. To track dynamic environment, the points with best exact fitness in each cluster are added into new cradle swarm, if environmental change has occurred. A variety of experiments have been conducted on the moving peaks benchmark (MPB) with 500 change frequency in each environment. The experimental results have demonstrated that the proposed approach has a good performance.

Text summarization aims to compress a textual document to a short summary while keeping salient information.Extractive approaches are widely used in text summarization because of their fluency and efficiency.However, most of existing extractive models hardly capture intersentence relationships, particularly in long documents.They also often ignore the effect of topical information on capturing important contents.To address these issues, this paper proposes a graph neural network (GNN)-based extractive summarization model, enabling to capture intersentence relationships efficiently via graph-structured document representation.Moreover, our model integrates a joint neural topic model (NTM) to discover latent topics, which can provide document-level features for sentence selection.The experimental results demonstrate that our model not only substantially achieves state-of-the-art results on CNN/DM and NYT datasets but also considerably outperforms existing approaches on scientific paper datasets consisting of much longer documents, indicating its better robustness in document genres and lengths.Further discussions show that topical information can help the model preselect salient contents from an entire document, which interprets its effectiveness in long document summarization.

Although the proper use of idioms can enhance the elegance of writing, the active use of various expressions is a challenge because remembering idioms is difficult.In this study, we address the problem of idiom recommendation by leveraging a neural machine translation framework, in which we suppose that idioms are written in one pseudo target language.Two types of reallife datasets are collected to support this study.Experimental results show that the proposed approach achieves promising performance compared with other baseline methods.

User experience is essential for human-computer dialogue systems. However, it is impractical to ask users to provide explicit feedbacks when the agents’ responses displease them. Therefore, in this paper, we explore to predict users’ imminent dissatisfactions caused by intelligent agents by analysing the existing utterances in the dialogue sessions. To our knowledge, this is the first work focusing on this task. Several possible factors that trigger negative emotions are modelled. A relation sequence model (RSM) is proposed to encode the sequence of appropriateness of current response with respect to the earlier utterances. The experimental results show that the proposed structure is effective in modelling emotional risk (possibility of negative feedback) than existing conversation modelling approaches. Besides, strategies of obtaining distance supervision data for pre-training are also discussed in this work. Balanced sampling with respect to the last response in the distance supervision data are shown to be reliable for data augmentation.

Writing academic articles in English is a challenging task for non-native speakers, as more effort has to be spent to enhance their language expressions. This paper presents an academic writing assistant system called Write-righter, which can provide real-time hint and recommendation by analyzing the input context. To achieve this goal, some novel strategies, e.g., semantic extension based sentence retrieval and LDA based sentence structure identification have been proposed. Write-righter is expected to help people express their ideas correctly by recommending top N most possible expressions.

Xin Wang, Yuanchao Liu, Chengjie Sun, Baoxun Wang, Xiaolong Wang. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2015.

Domain-specific term extraction contributes to all domain-oriented natural language processing tasks. Given a small set of domain-specific terms as seed terms, new terms from unlabeled corpora can be extracted by bootstrapping a term classifier to discover the association between seed terms and new terms. Traditional term representation method for domain-specific term extraction represents a term in a feature space of documents, which depicts association of terms which share common documents. This representation can't depict the inner-document information of terms and requires extracted terms to occur in multiple documents. A new term representation method in global contextual space is proposed for domain-specific term extraction in this paper. This representation mechanism depicts the association of terms which share common global contexts. The information of terms within certain document and among corpora is depicted by global contexts. Experiments on Chinese web corpus show that the proposed domain-specific term extraction method with global contextual representation outperforms traditional method with representation mechanism in documents space. The improvement for low frequency terms is much higher for the proposed method.