31. Discovery of exact equations for integer sequencesBoštjan Gec, Sašo Džeroski, Ljupčo Todorovski, 2024, original scientific article Abstract: Equation discovery, also known as symbolic regression, is the field of machine learning that studies algorithms for discovering quantitative laws, expressed as closed-form equations or formulas, in collections of observed data. The latter is expected to come from measurements of physical systems and, therefore, noisy, moving the focus of equation discovery algorithms towards discovering approximate equations. These loosely match the noisy observed data, rendering them inappropriate for applications in mathematics. In this article, we introduce Diofantos, an algorithm for discovering equations in the ring of integers that exactly match the training data. Diofantos is based on a reformulation of the equation discovery task into the task of solving linear Diophantine equations. We empirically evaluate the performance of Diofantos on reconstructing known equations for more than 27,000 sequences from the online encyclopedia of integer sequences, OEIS. Diofantos successfully reconstructs more than 90% of these equations and clearly outperforms SINDy, a state-of-the-art method for discovering approximate equations, that achieves a reconstruction rate of less than 70%.
Keywords: symbolic regression, equation discovery, online encyclopedia of integer sequences
Published in DiRROS: 27.03.2025; Views: 117; Downloads: 48
 Full text (425,51 KB)
This document has many files! More... |
32. |
33. Probabilistic grammars for modeling dynamical systems from coarse, noisy, and partial dataNina Omejc, Boštjan Gec, Jure Brence, Ljupčo Todorovski, Sašo Džeroski, 2024, original scientific article Abstract: Ordinary differential equations (ODEs) are a widely used formalism for the mathematical modeling of dynamical systems, a task omnipresent in scientific domains. The paper introduces a novel method for inferring ODEs from data, which extends ProGED, a method for equation discovery that allows users to formalize domain-specific knowledge as probabilistic context-free grammars and use it for constraining the space of candidate equations. The extended method can discover ODEs from partial observations of dynamical systems, where only a subset of state variables can be observed. To evaluate the performance of the newly proposed method, we perform a systematic empirical comparison with alternative state-of-the-art methods for equation discovery and system identification from complete and partial observations. The comparison uses Dynobench, a set of ten dynamical systems that extends the standard Strogatz benchmark. We compare the ability of the considered methods to reconstruct the known ODEs from synthetic data simulated at different temporal resolutions. We also consider data with different levels of noise, i.e., signal-to-noise ratios. The improved ProGED compares favourably to state-of-the-art methods for inferring ODEs from data regarding reconstruction abilities and robustness to data coarseness, noise, and completeness.
Keywords: ordinary differential equations, equation discovery, mathematical modeling, system identification, symbolic regression, partial observability
Published in DiRROS: 27.03.2025; Views: 103; Downloads: 73
Full text (2,66 MB)
This document has many files! More... |
34. A filtering approach for applying the two-fluid model to gas-liquid flows on high resolution gridsBenjamin Krull, Richard Meller, Matej Tekavčič, Fabian Schlegel, 2024, original scientific article Abstract: The two-fluid model is usually combined with closure forces designed for applications on coarse grids, i.e. bubbles (or particles) are typically assumed to be smaller than a grid cell. Practical applications however include situations where the mesh is comparatively fine, e.g. when meshing the wall boundary layer or in cases with growing bubbles. This may lead to non-convergent behaviour in mesh studies or to void fraction oscillations. To tackle this problem, a filtering approach is proposed, based on an additional diffusion term in the continuity equation. This approach increases the robustness of the results in regions of high spatial resolution, significantly reducing mesh-dependency of the simulation results. The implementation is straightforward, without a need to solve any additional system of equations. It is analysed in four different bubbly flow cases with varying characteristics: 2D/3D, wedge, square, and cuboid computational domains, with resolutions up to 32 cells per bubble diameter, laminar and turbulent flows, and several ways of gas injection. The additional computational effort varies, but is moderate. The proposed approach is applicable in multi-field two-fluid models for which a stable Euler-Euler behaviour on fine meshes is required, for example to prepare the transfer to an interface-resolving volume-of-fluid representation in morphology-adaptive approaches.
Published in DiRROS: 27.03.2025; Views: 164; Downloads: 32
Full text (929,15 KB) |
35. Asymmetry propagation in a pipe flow downstream of a 90° sharp elbow bendBlaž Mikuž, Klemen Cerkovnik, Iztok Tiselj, 2024, original scientific article Abstract: Pipe bends disrupt the flow, resulting in an asymmetric velocity field across the pipe diameter (D). We examined the recovery length required for the flow to return to a symmetric velocity profile downstream of a sharp elbow. The wall-resolved Large Eddy Simulation (LES) approach was applied to reproduce turbulent fluid flow at Reynolds numbers (�� ) of 5600 and 10,000. An additional case in the transitional laminar-turbulent-laminar regime was analyzed at ��=1400 . This analysis explored the behavior of the Dean vortices downstream of the elbow and revealed that, in turbulent cases, these vortices reverse their vorticity direction in the region between 8 D and 10 D. However, they eventually decay in structure as far as 25 D from the elbow. Flow asymmetry was analyzed in a 100 D long pipe section downstream of the elbow using four different criteria: wall shear stress (WSS), streamwise velocity, its fluctuations, and vorticity fields. This study found that in turbulent flows, the distance required for flow recovery is a few tens of D and decreases with increasing ��. However, in the transitional case, the flow separation within the elbow induces instabilities that gradually diminish downstream, and flow asymmetry persists even longer than the 100 D length of our outlet pipe section. WSS proved sensitive for detecting asymmetry near walls, whereas flow profiles better revealed bulk asymmetry. It was also shown that asymmetry indicators derived from velocity fluctuations and vorticity were less sensitive than those obtained from streamwise velocity.
Keywords: cevno koleno, flow asymmetry, pipe elbows, secondary flow, flow separation, turbulent flow, recovery length, Large Eddy Simulation
Published in DiRROS: 27.03.2025; Views: 153; Downloads: 48
Full text (6,74 MB)
This document has many files! More... |
36. CCL2/C–C chemokine receptor type 2‐mediated interactions among mast cells, basophils, and endothelial cellsMaruša Rihar, Rajia Bahri, Vida Forstnerič, Silvia Bulfone‐Paus, Peter Korošec, 2025, original scientific article Abstract: Background IL-33 is involved in allergic processes by promoting the release of various mast cell (MC) chemokines, including CCL2. However, it is yet unclear which specific cell type is primarily responsible for producing CCL2 during acute allergic reactions. This study aims to investigate the role of IL-33 in promoting CCL2 production in mast cells and assess the effect of MC-derived CCL2 on basophil migration and endothelial permeability. Methods Human blood-derived MCs (hMCs) were generated from peripheral blood precursors, passively sensitized with IgE, treated with IL-33, and stimulated with anti-IgE. The concentrations of nine cytokines known to influence immune cell chemotaxis (CCL2, CCL5, CCL11, MIP-1α, IL-8, IL-10, IL-13, granulocyte-macrophage colony-stimulating factor (GM-CSF), and vascular endothelial growth factor (VEGF) were assessed in the supernatants of hMCs. Subsequently, we investigated the impact of MC-derived CCL2 on basophil migration in vitro, as well as its effect on endothelial monolayer permeability using human umbilical vein endothelial cells (HUVECs). Results Stimulation with anti-IgE induced a significant release of CCL2, GM-CSF, IL-8 and VEGF from hMCs. Additionally, incubation with IL-33 overnight increased the production of several cytokines. Mast cell-derived CCL2 not only enhanced basophil migration in vitro but also increased endothelial monolayer permeability in HUVECs. The effect was reversed by a C–C chemokine receptor type 2 (CCR2) antagonist, indicating the involvement of CCL2 signaling through the CCR2 receptor. Conclusions IL-33 induces the production of chemotactic cytokines in hMCs. Mast cell-derived CCL2 plays an important role in basophil chemotaxis in vitro and affects endothelial monolayer permeability in the HUVEC model.
Published in DiRROS: 27.03.2025; Views: 134; Downloads: 79
Full text (1,01 MB)
This document has many files! More... |
37. Extending intergranular normal-stress distributions using symmetries of linear-elastic polycrystalline materialsSamir El Shawish, 2024, original scientific article Abstract: Intergranular normal stresses (INS) are critical in the initiation and evolution of grain boundary damage in polycrystalline materials. To model the effects of such microstructural damage on a macroscopic scale, knowledge of INS is usually required statistically at each representative volume element subjected to various loading conditions. However, calculating INS distributions for different stress states can be cumbersome and time-consuming. This study proposes a new method to extend the existing INS distributions to arbitrary loading conditions using the symmetries of a polycrystalline material composed of randomly oriented linearelastic grains with arbitrary lattice symmetry. The method relies on a fact that INS distributions can be accurately reproduced from the first (typically) ten statistical moments, which depend trivially on just three stress invariants and a few material invariants due to assumed isotropy and material linearity of the polycrystalline model. While these material invariants are complex averages, they can be extracted numerically from a few existing INS distributions and tabulated for later use. Practically, only three such INS distributions at properly selected loadings are required to provide all relevant material invariants for the first 11 statistical moments, which can then be used to reconstruct the INS distribution for arbitrary loading conditions. The proposed approach is demonstrated to be accurate and feasible for an arbitrarily selected linear-elastic material under various loading conditions.
Keywords: linear-elastic material, INS distributions, polycrystalline materials
Published in DiRROS: 27.03.2025; Views: 124; Downloads: 59
Full text (1,56 MB)
This document has many files! More... |
38. |
39. T+cc via the plane wave approach and including diquark-antidiquark operatorsIvan Vujmilović, Sara Collins, Luka Leskovec, Emmanuel Ortiz Pacheco, M. Padmanath, Saša Prelovšek, 2024, published scientific conference contribution Keywords: tetraquark, lattice QCD studies, quark mass
Published in DiRROS: 27.03.2025; Views: 112; Downloads: 73
Full text (1,81 MB)
This document has many files! More... |
40. A short-term investigation of diel vertical migrations of the calycophoran Siphonophora in the open south Adriatic Sea (July 2003)Davor Lučić, Adam Benović, Mirna Batistić, Mira Morović, Ivona Onofri, Juan Carlos Molinero, Tjaša Kogovšek, Barbara Gangai, Marijana Miloslavić, 2011, original scientific article Abstract: Morning, midday, evening and night repetitive sampling was used to demonstrate calycophoran diel vertical migration (DVM) at a fixed station in the oligotrophic southern Adriatic Sea at 8 depth intervals. Of 20 calycophoran species collected, most of them inhabited layers of specific temperature and light intensity, and migrated during the night. The deep species Kephyes ovata is primarily non-migratory. Typical nocturnal movements to the surface were recorded for Lensia conoidea and sphaeronectes koellikeri. Lensia subtilis and eudoxoides spiralis migrated toward the surface in the evening but sank deeper at night. Lensia meteori and Lensia fowleri reached upper layers at midday. The DVM of sphaeronectes irregularis was irregular and independent of diel light patterns. Though low in numbers, some calycophorans were found above the thermocline at night. For mesopelagic species, however, higher summer surface temperatures proved an effective barrier to migration above 100 m. Thus, certain species could be represented as key species of planktonic cnidarian assemblages found within a certain depth range. Though not homogenous, and while some undertake nocturnal migrations in both directions, all are interrelated by preferences of bathymetric distribution and specific DVM behavior.
Keywords: sea, medusae, biological research, abundance, bathimetric distribution, siphonophora, calycophoran siphonophores, composition, Adriatic Sea
Published in DiRROS: 26.03.2025; Views: 74; Downloads: 44
Full text (444,66 KB)
This document has many files! More... |
