In the field of music generation using algorithms, processing time-series data has consistently been a complex task. To improve music generation with long sequences, insightful-unit-conditional variational autoencoder is proposed, which can enhance unit-conditional variational autoencoders with an improved attention mechanism. This model integrates TransformerXLs recurrent mechanism and relative positional encoding with measure-level granularity. For practical applications, a scheme is addressed that uses optical flow to extract motion features from video frames, quantifying motion rate and intensity. Furthermore, a dynamic correlation method is proposed to align video motion features with musical rhythm, guiding the model to generate melodies that match the videos rhythm.

Gas recognition technology has received considerable attention from researchers in recent years. While successful in controlled environments, these algorithms often struggle in real-world industrial applications due to two key challenges: the lack of standardized performance benchmarks across datasets and over-reliance on multiple sensors. The lack of benchmarks hinders effective model comparison, while extensive sensor requirements increase hardware complexity and costs, limiting deployment flexibility. To address these gaps, we propose four specialized evaluation metrics-–coverage, identification efficiency, confusion rate, and cross identification rate—that unify performance evaluation across diverse datasets and enable a more precise, fair assessment of gas recognition algorithms. Leveraging these metrics, we introduce Wavelet Attention Gated Recurrent Unit (GRU) (WAG), a novel model that reduces the sensor requirement while preserving competitive recognition accuracy. By integrating wavelet transformation with external attention for robust frequency-domain feature extraction with GRU structures for efficient sequence modelling, WAG substantially minimizes hardware complexity and cost, paving the way for scalable, real-world gas recognition applications in resource-constrained environments.

Large-scale multiobjective optimization problems (LSMOPs) continue to be challenging for existing multiobjective evolutionary algorithms (MOEAs). The main difficulties are that: 1) the diversity preservation in both the objective space and the decision space needs to be taken into account when solving LSMOPs and 2) the existing learning structures in current MOEAs usually make the learning operators only coincidentally serve convergence and diversity, leading to difficulties in balancing these two factors. Therefore, balancing convergence and diversity in current MOEAs is difficult. To address these issues, this article proposes a multiobjective particle swarm optimizer with enhanced balance of convergence and diversity (MPSO-EBCD). In MPSO-EBCD, a novel velocity update structure for multiobjective particle swarm optimization is put forward, dividing the convergence, and diversity preservation operations into independent components. Following the proposed update structure, a weighted convergence factor is introduced to serve the convergence strategy, whilst a diversity preservation strategy is built to uniformly distribute the particles in the searched space based on a proposed multidimensional local sparseness degree indicator. By this means, MPSO-EBCD is able to balance convergence and diversity with specific parameters in independent operators. Experimental results on LSMOP benchmarks and a voltage transformer optimization problem demonstrate the competitiveness of the proposed algorithm compared to several state-of-the-art MOEAs.

Despite the widespread use of Fourier transform (FT) networks and generative adversarial networks (GANs) in audio signal processing, their practical effectiveness in unsupervised offline systems has not yet reached a fully satisfying level. Accumulating substantial experience in recent years, this article showcases how to construct an optimized, efficient music generation system. In the proposed system, the short-time Fourier transform is employed to divide a long music signal into equally sized short melodic segments. Each short melodic segment undergoes FT, and a nonautoregressive parallel WaveGAN system is trained by jointly optimizing multiresolution spectrograms and adversarial loss functions. This approach effectively captures the time–frequency distribution of real music waveforms. In essence, the proposed music generation system is a self-feedback unsupervised model relying on specific melody and note model pruning techniques. To further refine the music evaluation mechanism, in addition to conducting data analysis on the output melodies, subjective evaluation mechanisms are also incorporated.

Summary The portfolio optimization problem with cardinality constraint is usually solved by exact algorithms, heuristic algorithms, or combinations of them. We decompose the cardinality constraint mean‐variance model, and determine the assets and proportions by double roulette wheel selection (DRWS) and quadratic programming (QP), respectively. Then the accuracy of the solution is improved by a local search after we obtain the preliminary solution by combining DRWS and QP. Experimental results show that the proposed algorithm achieves better accuracy and more efficiency than the algorithms in the literature. Therefore, we can see that the algorithm designed according to the characteristics of specific problems can improve the computational efficiency, but the algorithm needs to be adjusted for different problems.

Abstract Convergence and diversity are two crucial aspects for multi‐objective optimization. Over past few decades, many researchers have dedicated their efforts to design more efficient convergence mechanisms. Different from them, this article is focused on the further improvement of the diversity preservation. First, a recently proposed relative non‐dominance matrix is empirically analyzed. As a result, it is found that the relative non‐dominance matrix can be used as a diversity evaluator. Based on the relative non‐dominance matrix, this article proposed a new optimizer for multi‐objective optimization problems. Experimental results on two popular test suites and a portfolio optimization problem illustrate the superiority of the proposed optimizer over state‐of‐the‐art methods.

The balance of exploration and exploitation of particle swarm optimization (PSO) is still a great challenge in large scale optimization. In our previous work, an adaptive particle swarm optimizer with decoupled exploration and exploitation (APSODEE) has been proposed to solve this problem. However, it still shows room for further improvements. First, APSODEE guides particles with the best individual in each sub-swarm, which is adverse to swarm diversity preservation especially in case of large scale optimization. Second, APSODEE fails to lead particles moving to sparse areas which are also potentially promising. To address these issues, two modifications are incorporated into APSODEE including a partial updating strategy and a quality restrained local sparseness diversity measurement. The former is proposed to further enhance the algorithm's swarm diversity preservation ability while the latter is designed to help guiding updated particles moving towards the areas which are both sparse and potentially promising, resulting in the modified APSODEE. The experiments are conducted based on CEC 2013 benchmarks with 1000 dimensionality. The results show the competitiveness of the proposed algorithm.

Abstract As a crucial factor, population diversity greatly affects performances of swarm intelligence algorithms. Especially, for large‐scale optimization problems (LSOPs), the searching space is huge and the number of local optima dramatically increases. Hence to well address LSOPs, a healthy population diversity is helpful to prevent a swarm from premature convergence. However, this is a big challenge to balance exploration and exploitation for swarm intelligence algorithms. To handle with this issue, in this paper, we design a novel algorithm structure for swarm update. In the proposed algorithm, a swarm is divided into several groups and conduct competition in each group where the loser will learn from the winner and meanwhile the winner does nothing in the corresponding iteration. For diversity measurement, we abandon the distance‐based measurement, but employ a frequency‐based measurement, namely entropy indicator, so that the diversity maintenance can be conducted with a different measurement of convergence situation. In this way, the diversity maintenance and convergence can be conducted simultaneously and independently. The benchmarks on the suite of LSOPs are employed to validate the performance of a proposed algorithm. By comparing several state‐of‐the‐art competitor algorithms, the results demonstrate that the proposed algorithm is effective and competitive in dealing with LSOPs.

Summary Wireless sensor networks (WSN) have high value in the field of wireless communications. As the earliest WSN clustering protocol, Low Energy Adaptive Clustering Hierarchy (LEACH) can effectively reduce the energy consumption of data transmission in sensor networks. However, LEACH has some problems such as cluster head nodes are unevenly distributed. In this paper, a unified heuristic bat algorithm (UHBA) is proposed to optimize elections in cluster heads. This algorithm guarantees that the election of cluster heads can freely transform both global search and local search. Meanwhile, comparing with several other variants of the bat algorithm in CEC2013 test suite, it can be seen from results that UHBA has better performance. Moreover, the application of the algorithm on LEACH is better than other algorithms, which further proves that the algorithm has better results.

Particle Swarm Optimization (PSO) is a popular algorithm which is widely investigated and well implemented in many areas. However, the canonical PSO does not perform well in population diversity maintenance so that usually leads to a premature convergence or local optima. To address this issue, we propose a variant of PSO named Grouping PSO with Personal-Best-Position ($P_{best}$) Guidance (GPSO-PG) which maintains the population diversity by preserving the diversity of exemplars. On one hand, we adopt uniform random allocation strategy to assign particles into different groups and in each group the losers will learn from the winner. On the other hand, we employ personal historical best position of each particle in social learning rather than the current global best particle. In this way, the exemplars diversity increases and the effect from the global best particle is eliminated. We test the proposed algorithm to the benchmarks in CEC 2008 and CEC 2010, which concern the large scale optimization problems (LSOPs). By comparing several current peer algorithms, GPSO-PG exhibits a competitive performance to maintain population diversity and obtains a satisfactory performance to the problems.

Biogeography-based optimization (BBO) is a powerful evolutionary algorithm inspired from the science of biogeography. It mainly uses the biogeography-based migration operator to share the information among individuals. In canonical BBO, according to the principle of immigration and emigration, poor solutions are like to be completely replaced by better ones. Consequently, this will lead to reduction of the population diversity. On the other hand, for Particle Swarm Optimization, a particle will learn from the global best solution and its own history best solution, which also deteriorate population diversity. In this paper, Particle Swarm Optimization (PSO) employs the selection mechanism of BBO and provides its solution updating strategy for BBO. A good particle has a large probability to be learned, while a poor particle has a small probability to be learned. In this way, the whole swarm can eliminate the affects from only one solution. The simulation is done using fourteen benchmark functions, and the results demonstrate that this hybrid BBO-PSO algorithm works efficiently.

This paper presents an efficient and reliable approach based on swarm intelligence to solve the optimal power flow problem. The optimal setting of distributed generations (DGs) is addressed if one or more generators broken down in a power distribution system, to achieve minimization of fuel cost and voltage profile stability. The proposed approach employs particle swarm optimization (PSO) and group search optimizer (GSO) for optimal setting of DGs. These algorithms are executed and compared on IEEE 14-bus test system, respectively. The results confirm the effectiveness and potential application of the proposed swarm-based optimization method in power distribution systems.

Turbo Codes present a new direction for the channel encoding, especially since they were adopted for multiple norms of telecommunications, such as deeper communication, etc. To obtain an excellent performance, it is necessary to design robust turbo code interleaver and decoding algorithms. In this paper, we are investigating particle swarm algorithm as a promising optimization method to find good interleaver for the large frame sizes, as well as design the decoding optimization mode (PSO-Turbo); and apply the proposed PSO-Turbo codes mode to the security radio data transmission; in which, a kind of transport control proposal based on PSO-Turbo optimizer for CBTC wireless channel is designed and simulated to validate our method.