Deep neural-network based optimization for the design of a multi-element surface magnet for MRI applications

We present a combination of a CNN-based encoder with an analytical forward map for solving inverse problems. We call it an encoder-analytic (EA) hybrid model. It does not require a dedicated training dataset and can train itself from the connected forward map in a direct learning fashion. A separate regularization term is not required either, since the forward map also acts as a regularizer. As it is not a generalization model it does not suffer from overfitting. We further show that the model can be customized to either finding a specific target solution or one that follows a given heuristic. As an example, we apply this approach to the design of a multi-element surface magnet for low-field magnetic resonance imaging (MRI). We further show that the EA model can outperform the benchmark genetic algorithm model currently used for magnet design in MRI, obtaining almost 10 times better results.

Extremal-Micro Genetic Algorithm Model for Time-Cost Optimization with Optimal Labour Productivity

In a highly competitive manufacturing environment, it is critical to balance production time and cost simultaneously. Numerous attempts have been made to provide various solutions to strike a balance between these factors. However, more effort is still required to address these challenges in terms of labour productivity. This study proposes an integrated substitution and management improvement technique for enhancing the effectiveness of labour resources and equipment. Furthermore, in the context of time-cost optimization with optimal labour productivity, an extremal-micro genetic algorithm (Ex-μGA) model has been proposed. A real-world case from the labour-intensive medium-scale bus body fabricating industry is used to validate the proposed model performance. According to the results, the proposed model can optimize production time and cost by 34 % and 19 %, respectively, while maintaining optimal labour productivity. In addition, this study provides an alternative method for dealing with production parameter imbalances and assisting production managers in developing labour schedules more effectively.

An Evolutionary Fake News Detection Method for COVID-19 Pandemic Information

As the COVID-19 pandemic rapidly spreads across the world, regrettably, misinformation and fake news related to COVID-19 have also spread remarkably. Such misinformation has confused people. To be able to detect such COVID-19 misinformation, an effective detection method should be applied to obtain more accurate information. This will help people and researchers easily differentiate between true and fake news. The objective of this research was to introduce an enhanced evolutionary detection approach to obtain better results compared with the previous approaches. The proposed approach aimed to reduce the number of symmetrical features and obtain a high accuracy after implementing three wrapper feature selections for evolutionary classifications using particle swarm optimization (PSO), the genetic algorithm (GA), and the salp swarm algorithm (SSA). The experiments were conducted on one of the popular datasets called the Koirala dataset. Based on the obtained prediction results, the proposed model revealed an optimistic and superior predictability performance with a high accuracy (75.4%) and reduced the number of features to 303. In addition, by comparison with other state-of-the-art classifiers, our results showed that the proposed detection method with the genetic algorithm model outperformed other classifiers in the accuracy.

Use of Genetic Algorithms to Optimize Efficiency in Small-Scale Structures

The prevalence of algorithms and computational tools in the modern-day has intersected with nearly every field. Generative design, specifically those using genetic algorithms, is an increasingly effective, yet cost efficient way to generate architectural designs in modern engineering. Thus, we adopt a genetic algorithm model in pursuit of maximizing the durability of a structure when it is stressed while minimizing the material cost. After the model is formulated, the algorithm is able to approximate with high accuracy the load a small-scale structure is able to bear, as well as iterate upon its designs to maximize a fitness function.