Inference Model Research Articles

Bayesian inference of the spatial distribution of steel corrosion in reinforced concrete structures using corrosion-induced crack width

Observations of corrosion-induced crack widths offer crucial information about the corrosion states of steel reinforcements in reinforced concrete (RC) structures, enabling a cost-effective method for inferring corrosion states through inverse analysis. However, the uncertainty associated with the relationship between corrosion-induced cracking and steel weight loss necessitates a probabilistic inference method, especially when considering the spatial distributions of steel weight loss, which provides important information to estimate the load-bearing capacity loss of corroded RC structures. This paper proposes a Bayesian framework to infer the steel weight loss distribution in RC structures based on the observed corrosion-induced crack width. To reduce the dimensions of the Bayesian inference, a Karhunen-Loève transform is applied to extract the principal distribution features of the steel weight loss. The forward model of the Bayesian inference adopts a data-driven sequence-to-sequence transduction approach to predict corrosion-induced crack width from steel weight loss. This model incorporates a novel nonlinear convolution kernel for input encoding and a sparse polynomial chaos expansion for decoding, which proves more accurate and efficient than finite element simulations. The Hamiltonian Markov chain Monte Carlo (HMCMC) sampler is used to efficiently sample from the posterior distribution of the Bayesian inference. The case study of the proposed method demonstrated that Bayesian inference provides robust range estimation for the steel weight loss distribution, with its 95% confidence interval encompassing most observations. Additionally, the method efficiently inferred high-dimensional steel weight loss sequences up to 61 dimensions, taking advantage of the dimension reduction technique and the gradient-informed HMCMC sampler.

The application of the Bayesian linear regression model to optimize the maintenance of a programmable logic controller

In this paper, an optimization approach, which is based on the Bayesian Linear Inference (BLI) model, has been proposed for the maintenance of Programmable Logic Controllers (PLCs). The BLI model, which is implemented using historical data, incorporates maintenance indicators like the number of failures (NF), total downtime (TD), total unexpected intervals (TUI), mean time to repair (MTTR) and mean time between failures (MTBF). It offers a probabilistic framework for determining the influence of each predictor variable on PLC maintenance. The model produces posterior means, credible intervals, and standard deviations, which provide insights into the magnitude and uncertainty of these relationships. The results from the study show that factors like NF and TD are influenced by the magnitude and direction of the maintenance levels. Also, the R-squared score (0.85) also indicates how much of the variability in maintenance in the system. From the results obtained, the study can conclude that the BLI model can optimize PLC maintenance procedures by identifying essential components and their contributions. Also, it is able to estimate future maintenance requirements and helps with resource allocation and process optimization decisions.

Bayesian Estimation of Fixed Effects Models with Large Datasets*

AbstractIn hierarchical prior longitudinal models, random effects are estimated by the Gibbs sampler. We show that fixed effects can be handled by a similar Gibbs sampler under a diffuse prior on the unobserved heterogeneity. The dummy variable approach for fixed effects is computationally intensive and has the out‐of‐memory risk, while the Gibbs sampler can reproduce the dummy variable estimator without creating dummy variables, and therefore avoids the memory burden. Compared to alternating projections and other classical approaches, our method simplifies both inference and estimation of the limited dependent variable models with fixed effects. The proposed method is applied to a real‐world mortgage dataset for classification with three‐way fixed effects on banks, regions, and loan purposes.

Insertable Glucose Sensor Using a Compact and Cost-Effective Phosphorescence Lifetime Imager and Machine Learning.

Optical continuous glucose monitoring (CGM) systems are emerging for personalized glucose management owing to their lower cost and prolonged durability compared to conventional electrochemical CGMs. Here, we report a computational CGM system, which integrates a biocompatible phosphorescence-based insertable biosensor and a custom-designed phosphorescence lifetime imager (PLI). This compact and cost-effective PLI is designed to capture phosphorescence lifetime images of an insertable sensor through the skin, where the lifetime of the emitted phosphorescence signal is modulated by the local concentration of glucose. Because this phosphorescence signal has a very long lifetime compared to tissue autofluorescence or excitation leakage processes, it completely bypasses these noise sources by measuring the sensor emission over several tens of microseconds after the excitation light is turned off. The lifetime images acquired through the skin are processed by neural network-based models for misalignment-tolerant inference of glucose levels, accurately revealing normal, low (hypoglycemia) and high (hyperglycemia) concentration ranges. Using a 1 mm thick skin phantom mimicking the optical properties of human skin, we performed in vitro testing of the PLI using glucose-spiked samples, yielding 88.8% inference accuracy, also showing resilience to random and unknown misalignments within a lateral distance of ∼4.7 mm with respect to the position of the insertable sensor underneath the skin phantom. Furthermore, the PLI accurately identified larger lateral misalignments beyond 5 mm, prompting user intervention for realignment. The misalignment-resilient glucose concentration inference capability of this compact and cost-effective PLI makes it an appealing wearable diagnostics tool for real-time tracking of glucose and other biomarkers.

PA-Net: Trustworthy weakly supervised point cloud semantic segmentation with primary–auxiliary structure

Weakly supervised point cloud segmentation is a hot research topic with significant implications for practical applications such as autonomous driving. However, existing weakly supervised approaches primarily focus on improving prediction results while neglecting the reliability of the inference outcomes. To address this issue, this paper proposes a primary–auxiliary network PA-Net for point cloud segmentation and uncertainty estimation. PA-Net achieves excellent segmentation results and enables effective uncertainty assessment. Specifically, we first train PA-Net on sparsely annotated point cloud data, which allows our network to capture multi-scale information of labeled points and provide additional supervisory signals. Subsequently, we utilize the trained model for inference. We mainly conducted experiments on the indoor point cloud dataset S3DIS in a 1% labeled setting. PA-Net achieved a mean Intersection over Union (mIoU) of 64.5%, which surpasses most weakly-supervised methods. In particular, it exceeded PSD by 3.5% mIoU and SQN by 0.8% mIoU. Similarly, on the outdoor dataset Toronto3D, PA-Net also outperformed most weakly supervised methods. With only 0.01% labeling, it surpassed PSD by 11.99% mIoU and SQN by 20.34% mIoU. Lastly, to validate the effectiveness of uncertainty estimation, we visualized scene uncertainty, thereby demonstrating the practicality of our uncertainty estimation approach. Code will be made publicly available at https://github.com/NiuYingchun/PA-Net.

Visual information perception system of coal mine comprehensive excavation working face for edge computing terminal

AbstractAiming at the problems of low detection accuracy, high computational complexity and long‐time consumption of visual perception model in a complex mining environment, this research designs a visual information perception system of coal mine comprehensive excavation working face for an edge computing terminal. Firstly, the C3‐Fast feature extraction module, spatial pyramid pooling with cross‐stage partial connection (SPPCSPC) pooling module, bi‐directional feature pyramid network and lightweight decoupled detection head are used to optimize the YOLOv5s model, so as to construct the FSBD‐YOLOv5s multi‐object detection model. Secondly, the pruning and distillation algorithm is used to lighten the FSBD‐YOLOv5s model, and the model complexity is greatly reduced while maintaining the model detection accuracy. Further, the lightweight FSBD‐YOLOv5s model is migrated and deployed to the edge computing terminal platform and the TensorRT engine is used to accelerate model inference. Finally, experiments are carried out based on the data set of the coal mine comprehensive excavation working face. The experimental results show that on the edge computing terminal platform, the parameters and computational volume of the lightweight FSBD‐YOLOv5s model are reduced by 50.8% and 34.0%, while its detection accuracy and speed reach 94.0% and 43.7 fps, which can fully satisfy the requirements of the accuracy and real‐time for the coal mine engineering applications.

HGEE: Learning for Trajectory Prediction with Heterogeneous Graph Interaction and External Embedding of Unmanned Swarm Systems in Adversarial Environment

Trajectory prediction of unmanned swarm systems, serving as the foundation for behavioral and intentional cognition, has attracted extensive attention and made considerable progress in adversarial research. The influence of heterogeneous interaction relationships and external factors is crucial for trajectory prediction. Consequently, this paper proposes the Heterogeneous Graph with External Embedding (HGEE) network. We model the latent variables as multi-layer heterogeneous graphs based on prior knowledge of different interaction relationships and propose a method for calculating edge embeddings for heterogeneous graphs. Furthermore, we introduce a method that combines external environmental feature with historical observational trajectory data as the input for the decoder, enabling the model to learn the impacts of obstacles, targets, and desired formations on trajectories. We demonstrate that our approach surpasses state-of-the-art models in interaction inference and trajectory prediction through experiments on our proposed formation datasets based on consensus theory, across five evaluation metrics.

Correlates of chronic pain onset and recovery in the CoLaus cohort.

Only few previous cohort studies examined simultaneously predictors of chronic pain (CP) onset and recovery. Furthermore, these studies used various sociodemographic and pain-related characteristics, without standardized measures of sleep and depression. The present study aimed at expanding and strengthening these findings in a large Swiss population. We analysed data from a longitudinal cohort (n = 4602) collected at two time points separated by 5 years in Lausanne, Switzerland. We studied through two independent multivariable logistic regression models, the predictors of CP onset and recovery, including socio-demographic data as well as standardized measures of sleep and mood. Chronic pain was reported by 43.1% and 44.4% of participants, with 11.6% at the second follow-up reporting moderate or intense pain. Neuropathic pain, regardless of intensity, had a more negative impact on quality of life. An inferential model (n = 1331) identified the male sex as predictive for recovering from CP. Older age, being overweight or obese (compared to normal weight), higher depression scores and pain medication intake were predictive for sustained pain at the second follow-up. A second model (n = 1886) identified being overweight or obese (compared to normal weight), low quality of sleep and being a former smoker (compared to a non-smoker) as predictive for developing CP, while the male sex was lowering the risk. While sex and weight are associated with both recovery and new CP onset, separate variables also need to be considered in these processes, underlining specific factors to be addressed, depending on the context, whether preventive or therapeutic. Multivariable models in a Swiss cohort (N = 4602) associate male sex, not taking pain medication, normal weight, lower depression scores and younger age with recovery from chronic pain, while females, obese or overweight, having worse sleep and former smokers are associated with onset of new chronic pain. These common and separate factors need to be considered in treatment and prevention efforts.

Generalized fiducial inference for the GEV change-point model

Generalized extreme value (GEV) distribution is used to analyse the maximum from a block of data. It is very useful to describe the unusual event rather than the usual event. In this paper, we propose a change-point detection procedure for GEV distribution based on generalized fiducial inference. The fiducial distribution of the change-point location is constructed. Meanwhile, Markov Chain Monte Carlo method combined with Gibbs sampling and the Metropolis–Hastings algorithm is utilised to estimate the location of the change point and its confidence interval. In addition, the generalized fiducial factor is used to test whether there is a change point. Simulation results show that the proposed generalized fiducial method performs better in accuracy, robustness and the length of confidence intervals. Finally, we apply the proposed method to the annual maximum rainfall data in Beijing.

Blockchain-supported online banking scheme

Online banking not merely enables customers to make transactions within a short time and access transactional information anywhere but also facilitates financial institutions to simplify transaction procedures and reduce their operating expenses. Common concerns by both customers and financial institutions are the secure authentication measures for protecting sensitive financial information from cybersecurity threats and maintaining their trust in a secure environment. This paper presents a blockchain-supported scheme to establish secure authentication and mutual trust for handling online banking processes. The proposed solution’s model design, inference process, proof method, security analysis, and performance evaluation are included in this study. With the decentralized applications and smart contracts, that benefit from disruptive innovations in blockchain technology, the proposed scheme enables financial institutions to provide efficient, immutable, transparent, and secure online financial services for customers. In addition, the computational efforts of the proposed solution show that high performance can be achieved in executing financial transactions while safeguarding sensitive data and retaining trust in a decentralized environment which is some of the most significant weak security of the existing authentication schemes for online banking platforms.

Combining distribution modelling and phylogeography to understand present, past and future of an endangered spider

BackgroundUnderstanding how endangered species respond to climatic changes is fundamental for their conservation. Due to its restricted geographic range, its sensitivity to the ongoing global warming and its continuing decline, the Southwestern-Alpine endemic wolf spider Vesubia jugorum is currently classified as Endangered in the IUCN Red List. Here, we combined species distribution modelling (SDM) and phylogeographic inference to describe the present, the past and the future of this species in light of the mtDNA genetic structure of extant populations.ResultsPhylogenetic and network analyses show a high level of genetic differentiation and a strong genetic structure of the populations, likely explicable by a long history of isolation and survival in separate refugia. The SDM projection into past climatic conditions supports these results by showing a smaller distribution range compared to present, mostly restricted to the Maritime and Ligurian Alps, which possibly served as main refugium. Future forecast shows a significant shift in the bioclimatic range towards higher altitudes and latitudes, with a drastic decrease of habitat suitability in the central and south-eastern parts of the range, with consequent general loss of haplotype diversity.ConclusionSDM and phylogeographic inference support the hypothesis that the current distribution and the genetic structure of the extant populations mirror the survival in situ of Vesubia jugorum across repeated glacial and interglacial phases, in line with the ‘long-term stability hypothesis’. Future predictions show a significant shift in the bioclimatic range that V. jugorum will be likely unable to track, with profound impact on its long-term survival and its genetic diversity. Our considerations have implication for conservation genetics, highlighting the pivotal role of the transboundary protected areas of the SW-Alps in promoting conservation efforts for this species.

A note on Bayesian inference for the bivariate pseudo-exponential model

In this present work, we discuss the Bayesian inference for the Bivariate Pseudo-exponential distribution. Initially, we assume independent gamma priors and Pseudo-gamma priors for the Pseudo-exponential parameters. Possible conjugacy is investigated in both full and sub-models. In some special sub-cases, conjugate priors can be identified. We are primarily interested in deriving the posterior means for each prior assumed and comparing each of the posterior means with the maximum likelihood estimators. Finally, all the Bayesian analyses are illustrated with a simulation study to demonstrate the performance of Bayesian parameter estimates using a variety of informative and non-informative priors, also illustrated with real-life applications.

Bayesian-informed fatigue life prediction in shallow shell structures with the dual boundary element method

This study introduces an innovative approach for statistically inferring the Equivalent Initial Flaw Size Distribution (EIFSD) in shallow shell structures with the aim of predicting fatigue life. The proposed approach integrates Bayesian inference and surrogate modelling techniques, and utilizes a long crack growth model. The EIFSD is crucial for estimating fatigue life and ensuring structural durability. To showcase the proposed methodology, a numerical example featuring a fuselage window structure under multi-axial loading, employing the Dual Boundary Element Method (DBEM) for crack growth modelling, is presented. Efficiency gains in the inference of the EIFSD are achieved through the development of surrogate models, notably a Co-Kriging model that significantly reduces computational time and outperforms a Kriging model in error reduction. The study explores two assumptions for inspection data sourcing, leading to enhanced methodological effectiveness across varying uncertainty levels. It considers the separate effects of Paris Law constants and loading conditions under increased uncertainty, presenting cases that demonstrate the Co-Kriging model’s high accuracy and computational efficiency. The proposed methodology’s robustness is further validated by assessing the impact of prior distribution quality on Bayesian inference, showing the method’s adaptability to high uncertainty while maintaining a good level of accuracy. This study also demonstrates an approach for fatigue life estimation of structures using the inferred EIFSD. By evaluating the structure up to a critical crack length, the methodology allows for the accurate estimation of fatigue life, which, in turn, facilitates the determination of optimal inspection intervals. Notably, under conditions of high uncertainty, the method showcased a maximum difference of 1.51% in the estimated fatigue life when compared to the true EIFSD, underscoring the technique’s high accuracy and reliability.

The Influence of Parallel Computing on Building Deep Learning Model for the Classification of Bean Diseases

In recent years, the utilization of deep learning techniques for image classification has made significant strides in the field of agriculture. One of the key areas of interest in agriculture is the early detection and classification of diseases in crops, as this can have an insightful impact on crop revenue and quality. This research has investigated the influence of parallel computing on the performance of a deep learning-based classification model for diagnosing bean diseases. Specifically, we have explored the use of parallel computing frameworks to accelerate model training and inference, thereby enhancing the efficiency and effectiveness of disease classification. Our findings demonstrated the potential for parallel computing to accelerate model training. When training a bean disease classification model, we achieved an accuracy of 0.93 using parallel computing, compared to 0.83 with serial computing. Moreover, parallel computing significantly reduced training time, taking only 3 minutes compared to 51 minutes with serial computing.

Structured methods for parameter inference and uncertainty quantification for mechanistic models in the life sciences.

Parameter inference and uncertainty quantification are important steps when relating mathematical models to real-world observations and when estimating uncertainty in model predictions. However, methods for doing this can be computationally expensive, particularly when the number of unknown model parameters is large. The aim of this study is to develop and test an efficient profile likelihood-based method, which takes advantage of the structure of the mathematical model being used. We do this by identifying specific parameters that affect model output in a known way, such as a linear scaling. We illustrate the method by applying it to three toy models from different areas of the life sciences: (i) a predator-prey model from ecology; (ii) a compartment-based epidemic model from health sciences; and (iii) an advection-diffusion reaction model describing the transport of dissolved solutes from environmental science. We show that the new method produces results of comparable accuracy to existing profile likelihood methods but with substantially fewer evaluations of the forward model. We conclude that our method could provide a much more efficient approach to parameter inference for models where a structured approach is feasible. Computer code to apply the new method to user-supplied models and data is provided via a publicly accessible repository.

An Efficient Temperature Calibration Method Based on the Improved Infrared Forward Model and Bayesian Inference

An Efficient Temperature Calibration Method Based on the Improved Infrared Forward Model and Bayesian Inference

CATRO: Channel Pruning via Class-Aware Trace Ratio Optimization.

Deep convolutional neural networks are shown to be overkill with high parametric and computational redundancy in many application scenarios, and an increasing number of works have explored model pruning to obtain lightweight and efficient networks. However, most existing pruning approaches are driven by empirical heuristics and rarely consider the joint impact of channels, leading to unguaranteed and suboptimal performance. In this article, we propose a novel channel pruning method via c lass-aware t race r atio o ptimization (CATRO) to reduce the computational burden and accelerate the model inference. Utilizing class information from a few samples, CATRO measures the joint impact of multiple channels by feature space discriminations and consolidates the layerwise impact of preserved channels. By formulating channel pruning as a submodular set function maximization problem, CATRO solves it efficiently via a two-stage greedy iterative optimization procedure. More importantly, we present theoretical justifications on convergence of CATRO and performance of pruned networks. Experimental results demonstrate that CATRO achieves higher accuracy with similar computation cost or lower computation cost with similar accuracy than other state-of-the-art channel pruning algorithms. In addition, because of its class-aware property, CATRO is suitable to prune efficient networks adaptively for various classification subtasks, enhancing handy deployment and usage of deep networks in real-world applications.

Design and implementation of efficient distributed deep learning model inference architecture on serverless computation

Although the distributed algorithm and architecture have been widely used in the development of traditional large-scale machine model reasoning tasks, but it is unavoidable about the source waste caused by synchronization between learners and participants. However, serverless computing is charged based on usage, which makes it a popular alternative for model inference. This article mainly proposes the architecture and algorithm that can be used for the inference task of the distributed deep learning model and adopts coarse-grained parallelization strategy to solve the problem of high communication cost between Serverless platform functions. The paper tested the performance of the model on the deep learning model and found that the model reasoning based on the Serverless platform can better cope with the requirements and changes, and ensure the stability of the system. It is worth noting that both the data owner and the model owner can benefit from the process of carrying out the machine learning model inference task, but existing solutions cannot satisfied with them privacy requirements, which is a problem needs to think about in the field of machine learning inference tasksbased on the Serverless platform in the future.

Socio-economic and cultural drivers of local perceptions and willingness to consume edible insects in Benin

Insects serve as alternative protein sources for humans. While entomophagy, the practice of eating insects, has deep historical roots in some regions, it remains less common in Benin. Understanding perceptions and determinants of entomophagy can guide actions for promoting the consumption of edible insects. The present study aimed to assess local perceptions of consuming edible insects and the drivers of their adoption. An ethnozoological study using interviews was conducted with 450 individuals from rural and peri‑urban areas in Benin's humid and semi-arid zones. Data were analyzed using descriptive statistics, univariate inference, and classification models. Findings showed greater acceptance of insects’ consumption in the semi-arid zone (33.56 %) than in the humid zone (23.11 %) and in rural areas (33.11 %) than in peri‑urban areas (23.56 %). Edible insects’ consumption was also higher in rural areas (34.22 %) than in peri‑urban areas (27.11 %), with no significant difference between biogeographical zones. Gender, age, and biogeographical zones were key determinants of local perceptions regarding the consumption of edible insects. Furthermore, factors influencing the decision to adopt the consumption of edible insects included perceptions of their nutritional value, gender, duration of residence, and geographical location. These factors can inform awareness raising about the usefulness of edible insects in promoting their consumption.

Deep Learning Approach to Detect Fake Video on Raspberry PI

Abstract: This project proposes a deep learning-based approach for real-time detection of fake videos on a resource-constrained device, specifically the Raspberry Pi. The solution combines the power of computer vision and recurrent neural networks to discern manipulated content from authentic videos effectively. The methodology involves using a pre-trained ResNeXt model for feature extraction, capturing spatial information from each video frame. These features are then fed into a Long Short-Term Memory (LSTM) network, allowing the model to understand and exploit temporal dependencies within the sequence of frames. The LSTM network learns patterns and nuances indicative of authentic or manipulated video content. The training process involves a carefully curated dataset containing both real and fake videos. The model is fine-tuned to optimize its performance, and metrics such as accuracy, precision, recall, and F1 score are employed for evaluation. To accommodate the constraints of the Raspberry Pi, the model is further optimized through techniques such as quantization, ensuring a balance between model size and inference accuracy. The final model is deployed on the Raspberry Pi, with a user-friendly interface capturing video frames in real-time. The system provides instantaneous feedback, indicating whether the observed video content is genuine or manipulated. This project contributes to the growing field of deepfake detection while addressing the challenges of implementing sophisticated models on edge devices. The combination of ResNeXt and LSTM offers a robust solution for discerning manipulated videos, making it suitable for real-world applications where computational resources are limited.