Computational Effort Research Articles

Cheminformatic Identification of Tyrosyl-DNA Phosphodiesterase 1 (Tdp1) Inhibitors: A Comparative Study of SMILES-Based Supervised Machine Learning Models.

Tyrosyl-DNA phosphodiesterase 1 (Tdp1) repairs damages in DNA induced by abortive topoisomerase 1 activity; however, maintenance of genetic integrity may sustain cellular division of neoplastic cells. It follows that Tdp1-targeting chemical inhibitors could synergize well with existing chemotherapy drugs to deny cancer growth; therefore, identification of Tdp1 inhibitors may advance precision medicine in oncology. Current computational research efforts focus primarily on molecular docking simulations, though datasets involving three-dimensional molecular structures are often hard to curate and computationally expensive to store and process. We propose the use of simplified molecular input line entry system (SMILES) chemical representations to train supervised machine learning (ML) models, aiming to predict potential Tdp1 inhibitors. An open-sourced consensus dataset containing the inhibitory activity of numerous chemicals against Tdp1 was obtained from Kaggle. Various ML algorithms were trained, ranging from simple algorithms to ensemble methods and deep neural networks. For algorithms requiring numerical data, SMILES were converted to chemical descriptors using RDKit, an open-sourced Python cheminformatics library. Out of 13 optimized ML models with rigorously tuned hyperparameters, the random forest model gave the best results, yielding a receiver operating characteristics-area under curve of 0.7421, testing accuracy of 0.6815, sensitivity of 0.6444, specificity of 0.7156, precision of 0.6753, and F1 score of 0.6595. Ensemble methods, especially the bootstrap aggregation mechanism adopted by random forest, outperformed other ML algorithms in classifying Tdp1 inhibitors from non-inhibitors using SMILES. The discovery of Tdp1 inhibitors could unlock more treatment regimens for cancer patients, allowing for therapies tailored to the patient's condition.

Optimizing indoor air models through k-means clustering of nanoparticle size distribution data

Sectional physics-based aerosol models imply a computational effort that hinders their use in building digital twins, real-time predictive control, and computer-based iterative optimization versus black-box approaches. The innovation of this paper lies in the proposal of a novel systematic methodology to optimize the number of size bins in sectional reduced-order models for particle concentration simulations. This allows its application in indoor air quality management and overcomes generalizability and data-dependency issues of black-box models. This method, based on k-means clustering, aims to ensure precision when tackling relatively fast fine and ultrafine particle simulations targeting the reduction of time and resource consumption from experimentally-determined aerosol size distribution time series. Consequently, three tests were carried out using combustion aerosols inside a custom-designed emission chamber to simulate emission hotspots in non-commercial and occupational settings. 2-, 3-, 4-, and 5-cluster classifications were evaluated for data coming from 13 particle size bins through silhouette analysis and the study of their temporal profiles. Results show that the 4-cluster classification summarizes the behavior of data in the 10–420 nm range, ensuing up to a 77 % improvement in the model's computational demand. Moreover, this method allows an accurate definition of the necessary size ranges to calculate nanoparticle concentrations inside the chamber and facilitates the interpretation of aerosol behavior and processes through the resulting clusters' temporal profiles.

Computationally effective machine learning approach for modular thermal energy storage design

This research presents an innovative approach that integrates computational fluid dynamics (CFD) and machine learning (ML) for the design and optimization of thermal energy storage (TES) systems. Heat discharging parametric analyses conducted using CFD serve as the basis for training ML models, including linear regression, K-nearest neighbor (KNN) regression, gradient boost regression (GBR), XGBoost, LightGBM, and neural network (NN). NN emerges as the most suitable for predicting time-dependent variations of concrete and heat transfer fluid (HTF) temperatures. The trained ML models offer an efficient alternative to traditional CFD simulations, enabling the prediction of temperatures in concrete thermal energy storage (CTES) modules under varying inlet conditions, velocities, and time. Leveraging these ML models, the research demonstrates the design of modular CTES cascaded systems with multiple modules in series and parallel configurations, significantly reducing computational cost and time by over 99% compared to full-scale CFD simulations. For instance, in predicting 4-hour time-dependent thermal behavior, CFD takes 97 s per data point and 238,500 s for a single module, compared to ML models’ 16-20 ms per data point and around 290 s per module, indicating their efficiency and scalability in predicting thermal discharge, especially for modular CTES system design and optimization. ML models also demonstrate computational efficiency for designing CTES systems involving multiple modules, taking approximately 765 s - 1047 s for various CTES system configurations, indicating their effectiveness over CFD in predicting thermal discharge for modular CTES systems. The integration of CFD and ML provides a streamlined workflow for designing and optimizing CTES systems, reducing computational efforts, cost, and time. Moreover, this workflow can be updated with additional training data to implement it for unique modular designs with different conditions. Such a generalization of this ML-based approach makes it applicable to a wide range of thermal energy storage designs and geometries, offering a promising avenue for future research and development in the field of thermal energy storage.

Performance analysis of the FB-PLL applied to real time monitoring of deep sleep brainwaves

Repose is indispensable for the human body, one of its main functions being to replenish energy. During sleep, the body's important restorative activities are performed by the organism, which is a basic human need. With awareness of the dimension that sleep represents and the need to improve its quality, the importance of monitoring in real time becomes evident. Based on this, the fast fourier transform is a technique capable of obtaining characteristics of a signal, such as its frequency, as well as being able to be applied to sleep waves. However, to have a good use, a considerable number of sample experimental cycles is necessary, which demands a certain computational effort. In this sense, this work we sought to apply the Phase-Locked Loop technique. Thus, a Fourier-based synchronization algorithm is proposed to track low-frequency brain waves during sleep. It is also worth noting that this method is capable of performing the estimates in just one cycle. Therefore, the objective is to obtain the estimated curve referring to the frequency and an efficient synchronization algorithm that is able to track deep sleep in real time.

A Human-in-the-Loop Method for Annotation of Events in Biomedical Signals.

Building large-scale data bases of biomedical signal recordings for training artificial-intelligence systems involves substantial human effort in data processing and annotation. In the case of event detection, experts need to exhaustively scroll through the recordings and highlight events of interest. We propose an iterative annotation support algorithm with a human in the loop to improve the efficiency of the annotation process. Our algorithm generates proposal events based on an event detection model trained on incomplete annotations. The human only needs to verify candidate events proposed by the tool instead of scrolling through the entire data set. Our algorithm iterates between proposal generation and verification to leverage the human-in-the-loop feedback to obtain a growing set of event annotations. Our algorithm finds a substantial amount of events at a fraction of the human time spent when comparing with a benchmark method and the normal manual process, finding all events in one data set and 70% of events in another with the human-in-the-loop only viewing 20% of the data. Our results show that combining human and computer effort can substantially speed up the annotation process for events in biomedical signal processing. Due to its simplicity and minimal reliance on task-specific information, our algorithm is broadly applicable, unlocking substantial improvements in the scalability and efficiency of biomedical signal annotation.

GIC-Flow: Appearance flow estimation via global information correlation for virtual try-on under large deformation

The primary aim of image-based virtual try-on is to seamlessly deform the target garment image to align with the human body. Owing to the inherent non-rigid nature of garments, current methods prioritise flexible deformation through appearance flow with high degrees of freedom. However, existing appearance flow estimation methods solely focus on the correlation of local feature information. While this strategy successfully avoids the extensive computational effort associated with the direct computation of the global information correlation of feature maps, it leads to challenges in garments adapting to large deformation scenarios. To overcome these limitations, we propose the GIC-Flow framework, which obtains appearance flow by calculating the global information correlation while reducing computational regression. Specifically, our proposed global streak information matching module is designed to decompose the appearance flow into horizontal and vertical vectors, effectively propagating global information in both directions. This innovative approach considerably diminishes computational requirements, contributing to an enhanced and efficient process. In addition, to ensure the accurate deformation of local texture in garments, we propose the local aggregate information matching module to aggregate information from the nearest neighbours before computing the global correlation and to enhance weak semantic information. Comprehensive experiments conducted using our method on the VITON and VITON-HD datasets show that GIC-Flow outperforms existing state-of-the-art algorithms, particularly in cases involving complex garment deformation.

Reducing the computational effort of symbolic supervisor synthesis

Reducing the computational effort of symbolic supervisor synthesis

Towards the Automation of Non-destructive Fault Recognition: Enhancement of Robustness and Accuracy Through AI Powered Acoustic and Thermal Signal Analysis in Time, Frequency- and Time-Frequency Domains

AbstractNon-destructive inspection and analysis techniques are crucial for quality assessment and defect analysis in various industries. They enable for screening and monitoring of parts and products without alteration or impact, facilitating the exploration of material interactions and defect formation. With increasing complexity in microelectronic technologies, high reliability, robustness and thus, successful failure analysis is essential. Machine learning (ML) approaches have been developed and evaluated for the analysis of acoustic echo signals and time-resolved thermal responses for assessing their ability for defect detection. In the present paper different ML architectures were evaluated, including 1D and 2D convolutional neural networks (CNNs) after transforming time domain data into the spectral- and wavelet domains. Results showed that 2D CNNs processing data in wavelet domain representation performed best, however at the expense of additional computational effort. Furthermore, ML-based analysis was explored for lock-in thermography to detect and locate defects in the axial dimension based on thermal emissions. While promising, further research is needed to fully realize its potential.

GDPR-anonymous population overlap estimation

The feasibility of many collaborative, data-driven research projects in Public Health and Social Sciences depends on the size of the overlap between the populations of contributing data partners, that is, the number of individuals for whom all parties can provide data. Where the identity of respective population members is subject to privacy protections, the overlap is traditionally estimated using de-identified cryptographic tokens to represent individuals. This does however not meet the strict standard of anonymity under GDPR and potentially runs counter to emerging privacy laws of US States, and may thus impose significant regulatory and governance burdens on research institutions, even at the early scoping stage. We present a novel method for overlap estimation, where two data partners construct a single, anonymous Bloom filter for their populations from which the overlap can be estimated. Two new techniques -- (a) the randomization of the contribution of individual population members to the Bloom filter and (b) an interactive bit-flipping protocol between data partners -- are presented which ensure that the Bloom filters are safe against dictionary attack by one party against the other party. We discuss the security aspects of this method, present the required mathematical formulas for calculating the overlap estimate, and report on the practical performance in terms of computational effort and resulting accuracy. The goal of the method is to facilitate new research by reducing the cost and complexity of initial planning and by making the overlap estimate GDPR-anonymous, and thus safer and subject to less onerous compliance requirements.

Bayesian time‐lapse full waveform inversion using Hamiltonian Monte Carlo

AbstractTime‐lapse images carry out important information about dynamic changes in Earth's interior, which can be inferred using different full waveform inversion schemes. The estimation process is performed by manipulating more than one seismic dataset, associated with the baseline and monitors surveys. The time‐lapse variations can be so minute and localized that quantifying the uncertainties becomes fundamental to assessing the reliability of the results. The Bayesian formulation of the full waveform inversion problem naturally provides confidence levels in the solution, but evaluating the uncertainty of time‐lapse seismic inversion remains a challenge due to the ill‐posedness and high dimensionality of the problem. The Hamiltonian Monte Carlo can effectively sample over high‐dimensional distributions with affordable computational efforts. In this context, we explore the sequential approach in a Bayesian fashion for time‐lapse full waveform inversion using the Hamiltonian Monte Carlo method. The idea relies on integrating the baseline survey information as prior knowledge to the monitor estimation. We compare this methodology with a parallel scheme in perfect and a simple perturbed acquisition geometry scenario considering the Marmousi and a typical Brazilian pre‐salt velocity model. We also investigate the correlation effect between baseline and monitor samples on the propagated uncertainties. The results show that samples between different surveys are weakly correlated in the sequential case, while the parallel strategy provides time‐lapse images with lower dispersion. Our findings demonstrate that both methodologies are robust in providing uncertainties even in non‐repeatable scenarios.

An enhanced traceable access control scheme based on multi-authority CP-ABE for cloud-assisted e-health system

Due to the inherent openness of the Internet of Things (IoT), e-health systems that utilize this technology on a broad scale are susceptible to network attacks. The propose black and white box traceable multi-authority ciphertext policy attribute-based encryption scheme (BWMABE) that combines black and white box accountability with ciphertext policy attribute-based encryption (CP-ABE) to improve data security in e-health systems. The proposed BWMABE scheme enhances security by enabling traceable permissions leakage and employs distributed attribute and key management across multiple authorities. By outsourcing the decryption process to cloud service providers, the proposed BWMABE significantly reduces the computational burden on end-users, making it suitable for lightweight IoT devices with limited computing power. Compared to previous traceable CP-ABE methods, the proposed BWMABE requires less computational effort while supporting both black box and white box traceability. By means of thorough security analysis and performance assessment, the proposed BWMABE has been validated to provide robust security and high efficiency, demonstrating its effectiveness in protecting e-health systems.

Plain Capping for Improved Accuracy of Approximate One- and Two-Electron Densities at Two-Particle Coalescence Points.

The values of the one-electron and intracule densities at two-particle coalescence points that enter the expressions for relativistic corrections to energies of Coulombic systems cannot be efficiently computed with sufficient accuracy from approximate wave functions expressed in terms of cuspless basis functions such as the explicitly correlated Gaussians. A new approach to alleviation of this problem, called plain capping, is proposed. Unlike those offered by the previously published formalisms, such as the expectation value identities and integral transforms, the accuracy improvements effected by the plain capping are attained with negligible computational effort and minimum programming. In the case of the on-top two-electron densities, whose accurate computation is particularly costly, the plain capping constitutes the only viable means of error reduction available at present.

Constitutive model and microstructure evolution of thermal deformation behavior of in situ TiB2p/Al-Zn-Mg-Cu composites with high Zn content

The deformation behavior of in situ TiB2 particle reinforced Al-Zn-Mg-Cu matrix composites with high Zn content was investigated through thermal compression tests. A comparative analysis of several constitutive models was performed, including the phenomenological modified Johnson-Cook (JC) and strain-compensated Arrhenius-type models, as well as the physically based modified and optimized Zerilli-Armstrong (ZA) models. In contrast to other materials, it is found that the accuracy of the modified JC and ZA models is significantly affected by the reference strain rate. When this was factored in, the optimized ZA model demonstrated superior accuracy, with a correlation coefficient (R) of 0.9986 and an average absolute relative error (AARE) of 2.15 %, while requiring the fewest calculation efforts amongst all of the models attempted. The mechanisms of the microstructural evolution in the composites after thermal deformation were also investigated. This work advances the understanding of TiB2p/Al-Zn-Mg-Cu composites and introduces an improved modeling framework, which is expected to benefit future material design and performance.

Deep learning-based source term estimation of hydrogen leakages from a hydrogen fueled gas turbine

Hydrogen fueled gas turbine is an efficient and environmentally friendly energy conversion equipment. However, it is prone to gas leakages from corrosion-prone components and pipe connections. In order to address gas leakage problems, source term estimation (STE) can be applied to estimate the location and intensity of the leakage sources. Traditional STE methods are based on an optimization procedure using an atmospheric transport and dispersion model, which requires considerable computation efforts and cannot meet the need of real-time implementation. The complex flow field around the gas turbine, the possible existence of multiple leakage sources, and the high-dimensional time series data further increase the difficulty of the STE problem. To address these challenges, in the present study, a deep learning based STE approach is proposed. The basic idea is to use long short-term memory auto-encoder (LSTM-AE) network to extract the encoded features of multi-sensor data and then use a deep neural network (NN) to establish the relationship between the encoded features and the source term parameters of hydrogen leakages. The multi-sensor data are generated using computational fluid dynamics (CFD) analysis to simulate relevant hydrogen leakage scenarios of concern. The results show that compared with other machine learning algorithms, the proposed approach has an overall best performance in terms of the STE accuracy. Specifically, its hydrogen leakage source localization accuracy is 0.9798, and the leakage strength estimation R-squared (R2) can reach 0.9632. When the training data proportion is only 20–30%, the proposed approach can still perform well, indicating the ability of the model to effectively predict the location and strength of the leakage source even with relatively less training data.

Identifying interaction boundary of inverter-based generation in assessing system strength of power grids using relative electrical distance concept

The increasing use of inverter-based generation (IBG) in power grids raises concern about system strength. This is partly due to the inherent interactions among multiple IBGs in close proximity to one another. This paper proposes an approach to identifying the existential boundary of interaction in a network using the relative electrical distance (RED) concept. The mathematical formulation of the RED concept to address the interaction problem among the IBGs involved utilising the power system network’s admittance matrix to capture its structural characteristics. An interaction matrix derived from the RED values of all IBG pairs was then developed to identify the interacting IBG groups. The proposed approach was demonstrated using the IEEE 39-bus system and a practical 72-bus Nigerian power grid. Results showed that RED values effectively group interacting IBGs, with values closer to 0 signifying higher interaction levels, values closer to 1 indicating lower interaction, and a value of 1 denoting no interaction. Time-domain simulations confirmed the accuracy of the approach, demonstrating that the effect of control interaction propagates proportionally to neighbouring IBGs based on RED values. However, fault currents can influence the impact of control interactions. This approach, which requires less computational effort, provides a quick identification tool for potential areas of concern based on the degree of interaction, enhancing the reliability of power grids with high IBG penetration.

Assessing the equivalent spring modelling method of CFS elements encased in ultra-lightweight concrete

An efficient finite element approach was recently developed to analyse encased cold-formed steel (CFS) structures. This new technique replaced encasing material with unidirectional springs, analogous to the Winkler foundation concept, to shorten the analysis time while ensuring accuracy and reliability in predicting the structural behaviour of encased CFS components. In this paper, the validity, and limitations of the simplified spring model to represent outstanding plates were assessed. The investigation demonstrated that the simplified spring model could effectively predict the ultimate load for a wide range of ultra-lightweight concrete moduli (50-250 MPa) with an acceptable error. The analysis indicated that plate elements initially in cross-section class 4 without encasing material become at least class 3, or better as a consequence of encasing. Previously reported experiments were used to evaluate the performance of the ESM. The analysis demonstrated that the ESM can accurately predict the local failure ultimate load of encased CFS sections with an acceptable error percent and significantly less computational effort than a 3D solid model.

A road map to cosmological parameter analysis with third-order shear statistics

Context. Third-order lensing statistics contain a wealth of cosmological information that is not captured by second-order statistics. However, the computational effort it takes to estimate such statistics in forthcoming stage IV surveys is prohibitively expensive. Aims. We derive and validate an efficient estimation procedure for the three-point correlation function (3PCF) of polar fields such as weak lensing shear. We then use our approach to measure the shear 3PCF and the third-order aperture mass statistics on the KiDS-1000 survey. Methods We constructed an efficient estimator for third-order shear statistics that builds on the multipole decomposition of the 3PCF. We then validated our estimator on mock ellipticity catalogs obtained from N-body simulations. Finally, we applied our estimator to the KiDS-1000 data and presented a measurement of the third-order aperture statistics in a tomographic setup. Results. Our estimator provides a speedup of a factor of ∼100–1000 compared to the state-of-the-art estimation procedures. It is also able to provide accurate measurements for squeezed and folded triangle configurations without additional computational effort. We report a significant detection of tomographic third-order aperture mass statistics in the KiDS-1000 data (S/N = 6.69). Conclusions. Our estimator will make it computationally feasible to measure third-order shear statistics in forthcoming stage IV surveys. Furthermore, it can be used to construct empirical covariance matrices for such statistics.

Optimizing column location in flat slab system using Particle Swarm Optimization vs Genetic Algorithms

Abstract One of the key architectural advantages of a concrete flat slab system is its flexibility in design. The absence of beams and columns allows for larger, open spaces with minimal obstructions, giving architects the freedom to create unique and innovative layouts. However, the location of the middle columns (one of the main factors affecting the construction cost) is commonly aligned based on experience which may entail an uneconomic design alternative. This paper aims to propose different machine learning optimization algorithms, namely, particle swarm optimization (PSO) and genetic algorithms (GA) to optimize the location of columns within a flat slab plan considering different serviceability requirements. Both optimization systems showed good performance for allocating the concrete columns, However, the GA algorithm showed better results in terms of maximum deflection with less computational effort.

A topological network connectivity design problem based on spectral analysis

How to improve network connectivity and which parts of the network are vulnerable are critical issues. We begin by defining an equal distribution problem, in which supplies are distributed equally to all nodes in the network. We then derive a topological network connectivity measure from the convergence speed, which is the second minimum eigenvalue of a Laplacian network matrix. Based on the equal distribution problem, we propose a method for identifying critical links for network connectivity using the derivative of the second minimum eigenvalue. Furthermore, we develop a network design problem that maximizes topological connectivity within a budget creating strengthening network links. The problem is convex programming, and the solution is global. Furthermore, it can be converted into an identical semidefinite programming problem, which requires less computational effort. Finally, we test the developed problems on road networks in the Japanese prefectures of Ishikawa and Toyama to determine their applicability and validity.

A Surrogate Model's Decision Tree Method Evaluation for Uncertainty Quantification on a Finite Element Structure via a Fuzzy-Random Approach

A novel additive manufacturing method (AM)) constructs a three-dimensional model from a computer-aided design by adding material layer by layer. This technique produces a lightweight end product with complex geometries and has gained recognition among industrial players. Nonetheless, the mechanical properties and geometry components are the uncertainties that prevail in its structures. An alternative approach using the Finite Element Method (FEM) to analyse these uncertainties demands extensive computational effort and time consumption. Therefore, a machine learning (ML) tool using the surrogate modelling technique offers an alternative way to provide and predict simulation outcomes. This study applies two surrogate modeling approaches, the decision tree (DT) and the Gaussian process regression (GPR) methods. Output data from a FEM simulation with uncertainty elements are obtained for the training purposes of the surrogate models. Both ML methods can predict simulation results with high precision. Both approaches obtained an excellent coefficient of determination value, R2 of 0.998, and Root Mean Square Error, RMSE of 0.012, successfully reducing time consumption and computational effort. The DT method shows better robustness when compared to the GPR method. A value change in the input parameter significantly impacts the surrogate model's prediction performance. An adequate quantity of data input for the training phase of both surrogate models exhibits the FEM results with the presence of uncertainty and robustness.