Search

To enhance solution accuracy and training efficiency in neural network approximation to partial differential equations, partitioned neural networks can be used as a solution surrogate instead of a single large and deep neural network defined on the whole problem domain. In such a partitioned neural network approach, suitable interface conditions or subdomain boundary conditions are combined to obtain a convergent approximate solution. However, there has been no rigorous study on the convergence and parallel computing enhancement on the partitioned neural network approach. In this paper, iterative algorithms are proposed to enhance parallel computation performance in the partitioned neural network approximation. Our iterative algorithms are based on classical additive Schwarz domain decomposition methods. For the proposed iterative algorithms, their convergence is analyzed under an error assumption on the local and coarse neural network solutions. Numerical results are also included to show the performance of the proposed iterative algorithms. [ABSTRACT FROM AUTHOR]

Neural network ensembling is a common and robust way to increase model efficiency. In this paper, we propose a new neural network ensemble algorithm based on Audibert's empirical star algorithm. We provide optimal theoretical minimax bound on the excess squared risk. Additionally, we empirically study this algorithm on regression and classification tasks and compare it to most popular ensembling methods. [ABSTRACT FROM AUTHOR]

• A novel water transparency calculation algorithm is introduced, employing deep neural network and image processing technology. This algorithm obviates the need for a water gauge, allowing the calculation of water transparency solely through the analysisof Secchi disk videos. Notably, it demonstrates heightened detection efficiency and success rates. • An enhanced Siamese tracker, termed SiamDCFF (Siamese tracker based on dual correlation feature fusion), is proposed for Secchi disk recognition. Distinguishing features of SiamDCFF include a dual correlation module and a feature fusion module. On the Secchi disk dataset, SiamDCFF achieves the highest success rate, outperforming other methods detailed in this paper. • The simplified 3D-ResNet [41] is leveraged to assess the state of the Secchi disk, transforming the critical position determination problem into an optimization challenge. The objective function involves maximizing the number of visible and invisible Secchi disk images on both sides of the critical position. This method facilitates adaptive critical position determination. • A novel Gated Recurrent Unit based on auto-regression (ARGRU) is proposed, taking into account the motion characteristics of the Secchi disk. ARGRU achieves a higher success detection rate for water transparency compared to the water gauge recognition method. Notably, ARGRU outperforms RNN [42], GRU, and LSTM [43] on the dataset presented in this paper. Water transparency is traditionally assessed through visual inspection using a lowered Secchi disk (SD) into the water, with the disappearance depth of the SD recorded as the measure of transparency. However, manual and visually-dependent frequency measurements of the SD render this process labor-intensive and time-consuming. This paper presents a comprehensive machine-vision-based algorithm designed for the automatic computation of water transparency using Secchi disk videos. The algorithm leverages multiple deep neural networks, including an enhanced Siamese tracking model, a 3D-ResNet, and an improved GRU network. Trained models directly identify the SD in the video, calculating water transparency by processing pixel information. Experimental results demonstrate the algorithm's efficacy in estimating water transparency across diverse natural environments, achieving commendable accuracy (MAE = 3.6 cm MSE = 21.5 cm RMSE = 4.6 cm), rapid processing speed (average 6.87 s), and robust stability. In comparison with the former water-gauge-based algorithm, our proposed algorithm exhibits heightened efficiency and a superior detection success rate. [ABSTRACT FROM AUTHOR]

Despite the existence of evidence-based rehabilitation strategies that address biomechanical deficits, the persistence of recurrent ankle problems in 70% of patients with acute ankle sprains highlights the unresolved nature of this issue. Artificial intelligence (AI) emerges as a promising tool to identify definitive predictors for ankle sprains. This paper aims to summarize the use of AI in investigating the ankle biomechanics of healthy and subjects with ankle sprains. Articles published between 2010 and 2023 were searched from five electronic databases. 59 papers were included for analysis with regards to: i). types of motion tested (functional vs. purposeful ankle movement); ii) types of biomechanical parameters measured (kinetic vs kinematic); iii) types of sensor systems used (lab-based vs field-based); and, iv) AI techniques used. Most studies (83.1%) examined biomechanics during functional motion. Single kinematic parameter, specifically ankle range of motion, could obtain accuracy up to 100% in identifying injury status. Wearable sensor exhibited high reliability for use in both laboratory and on-field/clinical settings. AI algorithms primarily utilized electromyography and joint angle information as input data. Support vector machine was the most used supervised learning algorithm (18.64%), while artificial neural network demonstrated the highest accuracy in eight studies. The potential for remote patient monitoring is evident with the adoption of field-based devices. Nevertheless, AI-based sensors are underutilized in detecting ankle motions at risk of sprain. We identify three key challenges: sensor designs, the controllability of AI models, and the integration of AI-sensor models, providing valuable insights for future research. [Display omitted] • Application of sensors and artificial intelligence to analyse ankle biomechanics. • Lab sensors predict parameters, paving the way for post-modern rehabilitation. • Field devices show potential for ankle injury prediction. • Focus on streamlined smart sensors and effective data management in the future. [ABSTRACT FROM AUTHOR]

• This paper suggests a grating Moiré fringe-based error correction approach that uses an artificial neural network (ANN) and the quartile method (QM). • The approach uses QM for data cleaning operations to eliminate outliers and creates the ANN model with a structure of fully connected layer-normalization layer-fully connected layer for regression training in order to achieve lightweight and stability. • The proposed QM-ANN error correction model's mean absolute error (MAE) can reach 3.3 nm once the QM data has been cleaned and the ANN model has been trained. Many interference factors of the complex and changeable working environment of the grating sensor are the main factors affecting the accuracy; a grating Moiré fringe error correction model based on the quartile method (QM) and artificial neural network (ANN) is proposed in this paper. Firstly, the grating sensor's signals in the working process are collected, and three kinds of interference signals (temperature, humidity and vibration) from the grating sensor working environment are collected simultaneously. Secondly, the QM is used to detect outliers from the collected data and delete data outliers directly. Subsequently, the processed data are input into the ANN model for training. The model is lightweight and stable, because it adopts the cascaded structure of full connection layer-normalization layer-full connection layer, which does not need deploy the complex convolution layer module etc. Finally, the corrected data are output after being processed by the QM-ANN error correction model. The experimental results show that the mean absolute error (MAE) of the proposed QM-ANN error correction model is better than QM-convolutional neural network (CNN), QM-long short-term memory (LSTM) and ANN, CNN, LSTM models without QM. It shows that the proposed model has better robustness and practicability. [ABSTRACT FROM AUTHOR]

• Formulation of a chance-constrained model for optimal power-water flow. • Recasting the model as a tractable problem using uncertain margins and convex relaxation. • Development of a novel iterative algorithm integrating two deep neural networks. • Achievement of a solution error under 0.03 % with a speedup over 200. The optimal power-water flow (OPWF) is a fundamental tool for operating integrated power and water networks. To address uncertainties in renewable energy and load demands, researchers have used chance-constrained models to investigate OPWF under uncertainty. However, existing work has limitations, such as oversimplified network representations, intractable problem formulations, and high computational burdens. This paper presents a chance-constrained OPWF formulation that incorporates detailed network constraints and stochastic uncertainties from wind speeds and electricity/water demands. By leveraging the concept of quantiles of stochastic variables and convexification techniques, the model is recast into a deterministic mixed-integer second-order cone programming (MISOCP) problem with uncertain margins. To efficiently solve this problem, we propose a novel iterative algorithm assisted by deep neural networks (DNNs). The algorithm integrates two DNNs: one designed to solve power and water flow equations, accelerating the probabilistic evaluation of uncertain margins; and the other to predict pipeline flows and refine variable ranges, facilitating a warm-start approach to the MISOCP model. Case studies conducted on the IEEE 123-node feeder coupled with a 67-node water distribution network validate the two DNNs and the overall iterative algorithm. The findings demonstrate that the proposed DNN-assisted approach effectively balances system security and economy. Compared to existing algorithms, the relative error of the optimization solution is less than 0.03%, with a speedup ratio exceeding 200. [ABSTRACT FROM AUTHOR]

As the deployment of artificial intelligence (AI) algorithms at edge devices becomes increasingly prevalent, enhancing the robustness and reliability of autonomous AI-based perception and decision systems is becoming as relevant as precision and performance, especially in applications areas considered safety-critical such as autonomous driving and aerospace. This paper delves into the robustness assessment in embedded Deep Neural Networks (DNNs), particularly focusing on the impact of parameter perturbations produced by single event upsets (SEUs) on convolutional neural networks (CNN) for image semantic segmentation. By scrutinizing the layer-by-layer and bit-by-bit sensitivity of various encoder–decoder models to soft errors, this study thoroughly investigates the vulnerability of segmentation DNNs to SEUs and evaluates the consequences of techniques like model pruning and parameter quantization on the robustness of compressed models aimed at embedded implementations. The findings offer valuable insights into the mechanisms underlying SEU-induced failures that allow for evaluating the robustness of DNNs once trained in advance. Moreover, based on the collected data, we propose a set of practical lightweight error mitigation techniques with no memory or computational cost suitable for resource-constrained deployments. The code used to perform the fault injection (FI) campaign is available at https://github.com/jonGuti13/TensorFI2 , while the code to implement proposed techniques is available at https://github.com/jonGuti13/parameterProtection. [ABSTRACT FROM AUTHOR]

In recent years, the application of deep neural networks in the field of image classification has become increasingly widespread. However, existing researches have exposed the inherent vulnerabilities and opacity of these network models, rendering them susceptible to carefully crafted adversarial attacks. This poses significant security risks to their practical deployment. Given that the majority of models in real-world applications operate in a black-box manner, current research on black-box attacks targeting classification models remains inadequate, facing significant challenges such as incomplete robustness evaluations, low success rates of black-box attacks, and excessive resource consumption during the attack process. To address these challenges, this paper innovatively proposes a black-box attack algorithm based on a cycle-consistent generative network. This algorithm directly generates adversarial perturbations through a generative network that integrates an attention mechanism, deeply learning the complex relationship between adversarial samples and clean sample data. It supports both targeted and untargeted attack modes. Experimental validation on the SAT dataset demonstrates that our method achieves an average white-box attack success rate exceeding 96% in classification tasks and an average success rate of up to 69.4% in black-box transfer attacks, exhibiting excellent transferability across different models. This research contributes significantly to enhancing the effectiveness of black-box attacks, reducing attack costs, and providing novel ideas and methods for enhancing the security of deep learning models. • We proposed an optimized cycle-consistent generative network for adversarial attacks, ultimately enhancing the success rate against other diverse black-box models. • Our algorithm offers support for both targeted and non-targeted attacks. Our targeted approach confers directionality to the generated perturbations, enabling purposeful attacks tailored to the attacker's intentions. • A transformation distribution module is incorporated into our model. We introduce a distance loss function to our overall loss computation. • Our approach enables migration attacks across different models, achieving a black-box attack success rate superior to current methods, exceeding 69.4%. [ABSTRACT FROM AUTHOR]

Even though the demand for utilizing geopolymers is growing, the need for current standard guidelines to regulate compliance to address the complexity of the mix design could be one of the major hurdles of utilizing geopolymers vastly in construction. There is no straightforward standard that addresses the complexity of the mix design of geopolymers. Thus, this work addresses main factors affecting the compressive strength of slag based geopolymers and provide a tool for predicting it. This article includes experimental work to evaluate the properties of slag-based geopolymer binders and the development of a model using Artificial Neural Network (ANN) algorithms for predicting the performance of these slag-based geopolymer binders. In this paper, we have utilized and developed ANN models for optimizing slag-based geopolymer mixes based on precursor materials' physiochemical properties and activation solutions constituents that can enhance performance compressive strength prediction in construction applications. • A high-accuracy DOE model was developed and used to characterize the Geopolymeric paste. • A developed ANN model was superior in forecasting the compressive strength of AAS systems. • The choice of the alkali activator has minimal impact on the compressive strength. • The optimal water-to-slag ratio is approximately 0.28. • The optimal alkali oxide concentration is 6 %. [ABSTRACT FROM AUTHOR]

Metastatic femoral tumors may lead to pathological fractures during daily activities. A CT-based finite element analysis of a patient's femurs was shown to assist orthopedic surgeons in making informed decisions about the risk of fracture and the need for a prophylactic fixation. Improving the accuracy of such analyses ruqires an automatic and accurate segmentation of the tumors and their automatic inclusion in the finite element model. We present herein a deep learning algorithm (nnU-Net) to automatically segment lytic tumors within the femur. A dataset consisting of fifty CT scans of patients with manually annotated femoral tumors was created. Forty of them, chosen randomly, were used for training the nnU-Net, while the remaining ten CT scans were used for testing. The deep learning model's performance was compared to two experienced radiologists. The proposed algorithm outperformed the current state-of-the-art solutions, achieving dice similarity scores of 0.67 and 0.68 on the test data when compared to two experienced radiologists, while the dice similarity score for inter-individual variability between the radiologists was 0.73. The automatic algorithm may segment lytic femoral tumors in CT scans as accurately as experienced radiologists with similar dice similarity scores. The influence of the realistic tumors inclusion in an autonomous finite element algorithm is presented in (Rachmil et al., "The Influence of Femoral Lytic Tumors Segmentation on Autonomous Finite Element Analyses", Clinical Biomechanics, 112, paper 106192, (2024)). • A deep learning algorithm for automatic segmentation of lytic femoral tumors is developed. • The deep learning dice similarity score of ∼0.67 is obtained, higher than any former study. • Inter-individual dice similarity score variability between two radiologists was 0.73. • The automatic segmentation of lytic tumors is as good as a radiologist. [ABSTRACT FROM AUTHOR]

Most image inpainting algorithms have problems such as fuzzy images, texture distortion and semantic inaccuracy, and the image inpainting effect is limited when processing photos with large missing sections and resolution levels. The paper proposes an effective image inpainting algorithm via partial multi-scale channel attention mechanism and deep neural networks to address the above phenomenon that existing image inpainting methods using deep learning modules have insufficient perception and representation capabilities for multi-scale features with high proportion of irregular defects. Initially, we used the Res-U-Net module as a generator. The U-Net-like backbone network topology can achieve the encoding and decoding stages of damaged images. Secondly, the residual network structure was built in the encoder and decoder to improve the ability of the proposed network to extract and display the features of the damaged images. Finally, the partial multi-scale channel attention module was inserted in the skip connection with the decoder to increase the efficiency of using the low-level features of the original images. The experimental results of the research can show that the proposed method outperforms state-of-the-art methods in terms of subjective visual perception and objective evaluation indicators on the CelebA , Places2 and Paris Street View datasets. • The model has improved the ability to extract and express multi-scale features. • The method had included partial multi-scale channel attention and residual network. • The backbone network structure can realize encoding and decoding stage. • The structure of residual networks had constructed to enhance the abilities. [ABSTRACT FROM AUTHOR]