In this study, we present a comprehensive set of experimental data aimed at uncovering the mechanisms and regularities governing the deformation behavior of composites reinforced with continuous carbon fibers (CF) based on thermoplastic polymers. This work describes data extraction techniques that can later be used to optimize the mechanical properties of such structures using neural network models. This paper examines the thermoplastic polymer polysulfone (PSU) of the Ultrason S 2010 brand, which was used as the matrix material for the composites, while high-strength Toray T700SC fibers were used as the reinforcing fibers. Composite samples in the form of rods with a diameter of 1 mm were obtained by impregnating the fibers with a solution of polysulfone in N-methyl-2-pyrrolidone, followed by solvent removal. The collected dataset contains more than 600 tensile test results, including load-strain diagrams for different test conditions, data on the failure mechanisms of the specimens, and SEM images of the specimen microstructure in cross and longitudinal sections. This dataset will be useful for ML model development.

Machine learning models have become firmly established across all scientific fields. Extracting features from data and making inferences based on them with neural network models often yields high accuracy; however, this approach has several drawbacks. Symbolic regression is a powerful technique for discovering analytical equations that describe data, providing interpretable and generalizable models capable of predicting unseen data. Symbolic regression methods have gained new momentum with the advancement of neural network technologies and offer several advantages, the main one being the interpretability of results. In this work, we examined the application of the deep symbolic regression algorithm SEGVAE to determine the properties of two-dimensional materials with defects. Comparing the results with state-of-the-art graph neural network-based methods shows comparable or, in some cases, even identical outcomes. We also discuss the applicability of this class of methods in natural sciences.

Patient status information and the course of their treatment can provide crucial insights for predicting intervention outcomes (long-term clinical results). We present the results of an analysis of a dataset of patients with bifurcation coronary artery lesions and compare various machine learning approaches, including a novel approach using the Kolmogorov-Arnold neural network (KAN), to address the task of classifying bifurcation coronary artery lesions. We conducted a comparative analysis of the trained models to evaluate their effectiveness. The study was based on a multicenter registry for the treatment of patients with bifurcation coronary artery lesions. In total, 1961 patients were included in the analysis. The main result of this paper is the analysis of the application of KAN and its comparison with the multilayer perceptron (MLP). We demonstrated that the KAN model outperforms traditional machine learning algorithms and the MLP, both in terms of the AUC-ROC metric (0.7127 and 0.5909, respectively) and classification accuracy.

In recent years, diffusion-based models have demonstrated exceptional performance in searching for simultaneously stable, unique, and novel (S.U.N.) crystalline materials. However, most of these models don't have the ability to change the number of atoms in the crystal during the generation process, which limits the variability of model sampling trajectories. In this paper, we demonstrate the severity of this restriction and introduce a simple yet powerful technique, mirage infusion, which enables diffusion models to change the state of the atoms that make up the crystal from existent to non-existent (mirage) and vice versa. We show that this technique improves model quality by up to x2.5 compared to the same model without this modification. The resulting model, Mirage Atom Diffusion (MiAD), is an equivariant joint diffusion model for de novo crystal generation that is capable of altering the number of atoms during the generation process. MiAD achieves an 8.2

Two-dimensional C60 carbon allotropes have gained much attention since their first synthesis in 2022, but many of their thermophysical and mechanical properties remain unreported in the literature. In this article, we performed a high-temperature molecular dynamics study of quasi-hexagonal (qHP) and quasi-tetragonal (qTP) C60 phases using the modern machine-learning interatomic potential GAP-20. We show that, contrary to previous calculations, at T>1200 K, both phases are unstable and decompose into individual C60 molecules. A low bending modulus indicates the possibility of nanoripple excitation at high temperatures, similar to those in graphene. We also demonstrate the crucial role of interatomic potential verification for MD analysis of previously unexplored carbon allotropes.

Quantum dots (QDs) are very attractive nanostructures from an application point of view due to their unique optical properties. Optical properties and valence band (VB) state character was numerically investigated with respect to the effects of nanostructure geometry and composition. Numerical simulation was carried out using the Luttinger–Kohn model adapted to the particular case of QDs in inverted pyramids. We present the source code of the 4-band Luttinger–Kohn model that can be used to model AlGaAs or InGaAs nanostructures. The work focuses on the optical properties of GaAs/AlGaAs [111] QDs and quantum dot molecules (QDMs). We examine the dependence of Ground State (GS) optical properties on the structural parameters and predict optimal parameters of the QD/QDM systems to achieve dynamic control of GS polarization by an applied electric field.

Artificial Intelligence (AI) permeates all areas of our lives. Even now, we all use AI algorithms in our daily activities, and medicine is no exception. The potential of AI technology is hard to overestimate; AI has already proven its effectiveness in many fields of science and technology. A vast number of methods have been proposed and are being implemented in various areas of medicine, including interventional cardiology. A hallmark of this discipline is the extensive use of visualization techniques not only for diagnosis but also for the treatment of patients with coronary heart disease. The implementation of instrumental AI will reduce costs, in a broad sense. In this article, we provide an overview of AI research in interventional cardiology, practical applications, as well as the problems hindering the widespread use of neural network technologies in interventional cardiology.

Deep learning (DL) methodologies have led to significant advancements in various domains, facilitating intricate data analysis and enhancing predictive accuracy and data generation quality through complex algorithms. In materials science, the extensive computational demands associated with high-throughput screening techniques such as density functional theory, coupled with limitations in laboratory production, present substantial challenges for material research. DL techniques are poised to alleviate these challenges by reducing the computational costs of simulating material properties and by generating novel materials with desired attributes. This comprehensive review document explores the current state of DL applications in materials design, with a particular emphasis on two-dimensional materials. The article encompasses an in-depth exploration of data-driven approaches in both forward and inverse design within the realm of materials science.

We investigate experimentally and theoretically the impact of valence band mixing and spectrum of confined states on the polarization of light emitted from or absorbed by GaAs/AlGaAs semiconductor quantum dots and quantum wires with tailored heterostructure potential. In particular, such nanostructures with parabolic-profile confinement potentials, realized by organometallic vapor phase epitaxy inside pyramidal pits, served as model systems for the study. Different degrees of linear polarization (DOLP) of emitted light, depending on the confinement potential profile, the specific excitonic transition, and the level of excitation, are observed. A theoretical model shows that, besides the impact of valence band mixing, the overlap of conduction and valence band wavefunctions as well as state occupation probability and broadening of transitions determine the DOLP. The conclusions are useful for the design of quantum light emitters with controlled polarization properties.

Quantum Dots are very attractive nanostructures from an application point of view due to their unique optical properties. Optical properties and Valence Band states character was numerically investigated from the effect of nanostructure geometry and composition. Numerical simulation was carried out using Luttinger Kohn model adapted to the particular use case of QDs in inverted pyramids. We present the source code of the 4 band Luttinger Kohn model that can be used to model AlGaAs or InGaAs nanostructures. Here we focus on the optical properties study of GaAs/AlGaAs [111] QDs and Quantum Dot Molecules (QDMs). We examine the dependence of Ground State (GS) optical properties on their structural parameters and predict optimal parameters of the QD and QDM systems to achieve the dynamic control of GS polarization by the applied electric field.

There are many problems in physics, biology, and other natural sciences in which symbolic regression can provide valuable insights and discover new laws of nature. Widespread deep neural networks do not provide interpretable solutions. Meanwhile, symbolic expressions give us a clear relation between observations and the target variable. However, at the moment, there is no dominant solution for the symbolic regression task, and we aim to reduce this gap with our algorithm. In this work, we propose a novel deep learning framework for symbolic expression generation via variational autoencoder (VAE). We suggest using a VAE to generate mathematical expressions, and our training strategy forces generated formulas to fit a given dataset. Our framework allows encoding apriori knowledge of the formulas into fast-check predicates that speed up the optimization process. We compare our method to modern symbolic regression benchmarks and show that our method outperforms the competitors under noisy conditions. The recovery rate of SEGVAE is 65% on the Ngyuen dataset with a noise level of 10%, which is better than the previously reported SOTA by 20%. We demonstrate that this value depends on the dataset and can be even higher.

Quasi-one-dimensional AlGaAs quantum wires (QWRs) with parabolic heterostructure profiles along their axis were fabricated using metallorganic vapor phase epitaxy (MOVPE) on patterned (111)B GaAs substrates. Tailoring of the confined electronic states via modification in the parabolic potential profile is demonstrated using model calculations and photoluminescence spectroscopy. These novel nanostructures are useful for studying the optical properties of systems with dimensionality between zero and one.