Alignment of each optical element at a synchrotron beamline takes days, even weeks, for each experiment costing valuable beam time. Evolutionary algorithms (EAs), efficient heuristic search methods based on Darwinian evolution, can be utilized for multi-objective optimization problems in different application areas. In this study, the flux and spot size of a synchrotron beam are optimized for two different experimental setups including optical elements such as lenses and mirrors. Calculations were carried out with the X-ray Tracer beamline simulator using swarm intelligence (SI) algorithms and for comparison the same setups were optimized with EAs. The EAs and SI algorithms used in this study for two different experimental setups are the Genetic Algorithm (GA), Non-dominated Sorting Genetic Algorithm II (NSGA-II), Particle Swarm Optimization (PSO) and Artificial Bee Colony (ABC). While one of the algorithms optimizes the lens position, the other focuses on optimizing the focal distances of Kirkpatrick–Baez mirrors. First, mono-objective evolutionary algorithms were used and the spot size or flux values checked separately. After comparison of mono-objective algorithms, the multi-objective evolutionary algorithm NSGA-II was run for both objectives – minimum spot size and maximum flux. Every algorithm configuration was run several times for Monte Carlo simulations since these processes generate random solutions and the simulator also produces solutions that are stochastic. The results show that the PSO algorithm gives the best values over all setups.

The world’s high-resolution images are supplied by a radar system named Synthetic Aperture Radar (SAR). Semantic SAR image segmentation proposes a computer-based solution to make segmentation tasks easier. When conducting scientific research, accessing freely available datasets and images with low noise levels is rare. However, SAR images can be accessed for free. We propose a novel process for labeling Sentinel-1 SAR radar images, which the European Space Agency (ESA) provides free of charge. This process involves denoising the images and using an automatically created dataset with pioneering deep neural networks to augment the results of the semantic segmentation task. In order to exhibit the power of our denoising process, we match the results of our newly created dataset with speckled noise and noise-free versions. Thus, we attained a mean intersection over union (mIoU) of 70.60% and overall pixel accuracy (PA) of 92.23 with the HRNet model. These deep learning segmentation methods were also assessed with the McNemar test. Our experiments on the newly created Sentinel-1 dataset establish that combining our pipeline with deep neural networks results in recognizable improvements in challenging semantic segmentation accuracy and mIoU values.

The escape rate of S → R switching (from superconducting S state to resistive R state) in Josephson junction formed by s-wave single-band (SB) and three-band (ThB) superconductors (SB/ThB junctions) in macroscopic quantum tunneling regime is investigated. We use the effective critical current approximation in SB/ThB junctions. It was shown that escape rate can exhibit qualitative features in the case of frustration effects in ThB superconductors. Inclusion of Leggett modes in ThB superconductors leads to enhancement of escape rate in three-channel Josephson junction in comparison with single-channel junctions.

Accurate prediction of the electronic band gap is essential for accelerating the discovery and design of semiconducting and energy materials. Conventional density functional theory (DFT) methods, while physically rigorous, remain computationally expensive and limited in scalability. In this study, we propose a hybrid artificial intelligence framework that combines graph-based deep learning embeddings with classical machine learning algorithms to achieve high-accuracy, interpretable, and computationally efficient band gap prediction. The model integrates embeddings obtained from CGCNN, MEGNet, and SchNet architectures with physically meaningful crystal descriptors─including maximum electronegativity, crystal system, space group, and spin–orbit coupling─and trains them using optimized gradient-boosting and neural architectures. Trained on 136,000 crystal structures from the Materials Project database, the hybrid model achieves R² = 0.921, MAE = 0.191, and MSE = 0.155, outperforming both classical models (Ward et al., 2016) and standalone graph neural networks such as CGCNN (Xie and Grossman, 2018). The achieved accuracy is statistically comparable to the state-of-the-art ALIGNN model (Choudhary et al., 2021), while requiring lower computational resources and offering enhanced generalization due to the integration of multisource structural information. SHAP-based interpretability analysis highlights that the model captures physically consistent relationships, with metallicity and magnetic site features emerging as dominant factors in band gap prediction. These findings demonstrate that the synergy between deep structural embeddings and classical algorithms provides a powerful, scalable approach for materials informatics. The proposed framework establishes a foundation for multiproperty prediction, transfer learning across databases, and inverse materials design driven by interpretable artificial intelligence.