Spotted Hyena Optimizer (SHO) is a population-based metaheuristic algorithm inspired by the spotted hyenas' social behavior, and it has been developed to solve global optimization problems. SHO shown superior performance over its competitive algorithms in solving benchmark function engineering design However, suffers from getting stuck local optima due lack of exploration while multi-modal This article proposes an improved SHO, quantum (QSHO), computing. The QSHO implements computing...

Convolutional neural networks (CNNs) are deep learning models used widely for solving various tasks like computer vision and speech recognition. CNNs developed manually based on problem-specific domain knowledge tricky settings, which laborious, time consuming, challenging. To solve these, our study develops an improved differential evolution of convolutional network (IDECNN) algorithm to design CNN layer architectures image classification. Variable-length encoding is utilized represent the...

Tunicate Swarm Algorithm (TSA) is a novel swarm intelligence algorithm developed in 2020. Though it has shown superior performance numerical benchmark function optimization and six engineering design problems over its competitive algorithms, still needs further improvements. This article proposes two improved TSA algorithms using chaos theory, opposition-based learning (OBL) Cauchy mutation. The proposed are termed OCSTA COCSTA. static dynamic OBL used respectively the initialization...

Recently, convolutional neural networks (CNNs) have shown promising achievements in various computer vision tasks. However, designing a CNN model architecture necessitates high-domain knowledge expert, which can be difficult for new researchers while solving real-world problems like medical image diagnosis. Neural search (NAS) is an approach to reduce the human intervention by automatically architecture. This study proposes two-phase evolutionary framework design suitable classification...

In today's AI-driven era, deep learning (DL) algorithms play a crucial role in automatically detecting life-threatening skin cancers, thereby significantly enhancing survival rates. It makes cancer detection using DL an exciting area of exploration. While much the prior research has focused on single-model approaches, combining ensembles multiple models can enhance classification accuracy. Previous studies mainly relied convolutional neural networks (DCNNs), which have limitations capturing...

Voice conversion (VC) emerged as a significant domain of research in the field speech synthesis recent years due to its emerging application voice-assistive technologies, such automated movie dubbing speech-to-singing conversion, name few. VC deals with vocal style one speaker another while keeping linguistic contents unchanged. Nowadays, generative adversarial network (GAN) models are widely used for feature mapping from source target speaker. In this article, we propose an...

Particle Swarm Optimization (PSO) has shown its good search ability in many optimization problem. But PSO easily gets trapped into local optima while dealing with complex problems. In this work, we proposed an improved PSO, namely PSO-APM, which adaptive polynomial mutation strategy is employed on global best particle the hope that it will help particles jump out optima. carried our experiments 8 well-known benchmark Finally results are compared classical and power (PMPSO).

Activation functions in the neural networks play an important role by introducing non-linear properties to networks. Thus it is considered as one of essential ingredients among other building blocks a network. But selection appropriate activation function for enhancement model accuracy strenuous sense; performance NN-model influenced proper dataset. Proper still trial and error method which improves classification. As solution this problem, we have proposed ensembling majority voting that...

Convolutional neural networks (CNNs) are broadly used to solve various computer vision tasks. However, designing optimal CNN architecture for solving a particular task is trial- and-error process and thus requires domain-specific expertise. Recently, several meta-heuristic strategies have been developed the design challenges of models. Here, we propose an approach using artificial bee colony (ABC) algorithm named as ABC-CNN. A variable-length encoding scheme refinement strategy proposed...

Voice conversion (VC) is the process of converting vocal texture a source speaker similar to that target without altering content speaker's speech. With ongoing developments deep generative models, adversarial networks (GANs) appeared as better alternative conventional statistical models for VC. However, existing VC model-generated speech samples possess substantial dissimilarity from their corresponding natural human Therefore, in this work GAN-based model proposed which incorporated with...

Drug is a very much essential substance in health care system. Producing new drug for disease the market using traditional method time consuming and expensive. Recently, design process sped-up by computer resources known as Computer-Aided Design (CADD). In design, bio-molecules are responsible to produce drug. Therefore, molecules identification an part of CADD. this paper, two machine learning algorithms such Naive Bayesian (NB) classifier k-Nearest Neighbors (k-NN) evaluated classify...