Input Variance Research Articles

PERFORMANCE-BASED PAYMENT ADJUSTMENT FACTOR FOR CONCRETE PAVEMENT

The increasing adoption of adjustable payment plans in road construction has prompted the exploration of a performance-based pay factor for concrete pavement, focusing on the discrepancy between designed and actual pavement performance due to variations in material and construction factors. This research proposes a rational methodology to adjust contractors' bid prices based on the impact of quality deviations on the pavement's total life cycle cost, including initial construction, maintenance, rehabilitation, and user costs. Utilizing performance modeling to predict pavement performance, the study introduces a formula for calculating payment adjustments that consider predicted lifespan, annual maintenance costs, and the real discount rate. The approach centers on the American Association of State Highway and Transportation Officials (AASHTO) rigid pavement thickness design equation, which integrates the variance of input and output factors in performance prediction. By focusing on controllable construction variables like concrete thickness and strength, the methodology aims to equate the annualized life cycle costs of designed and constructed pavements, thereby determining the payment adjustment. Simulation data supports the proposed payment adjustment factor, indicating its effectiveness in generating reasonable estimates for as-constructed pavement value. While the study acknowledges the benefits of incorporating the time value of money and a comprehensive life cycle cost analysis, it emphasizes the fundamental fairness of its method, which is based on expected pavement performance. Applied on a shadow specification basis to concrete paving projects in Texas, the research demonstrates the potential of tying material variability to actual performance, offering a foundational method for adjusting contractor payments based on expected pavement outcomes. (Abstract generated by AI tool ChatGPT 4)

Probabilistic 3D Correspondence Prediction from Sparse Unsegmented Images.

The study of physiology demonstrates that the form (shape) of anatomical structures dictates their functions, and analyzing the form of anatomies plays a crucial role in clinical research. Statistical shape modeling (SSM) is a widely used tool for quantitative analysis of forms of anatomies, aiding in characterizing and identifying differences within a population of subjects. Despite its utility, the conventional SSM construction pipeline is often complex and time-consuming. Additionally, reliance on linearity assumptions further limits the model from capturing clinically relevant variations. Recent advancements in deep learning solutions enable the direct inference of SSM from unsegmented medical images, streamlining the process and improving accessibility. However, the new methods of SSM from images do not adequately account for situations where the imaging data quality is poor or where only sparse information is available. Moreover, quantifying aleatoric uncertainty, which represents inherent data variability, is crucial in deploying deep learning for clinical tasks to ensure reliable model predictions and robust decision-making, especially in challenging imaging conditions. Therefore, we propose SPI-CorrNet, a unified model that predicts 3D correspondences from sparse imaging data. It leverages a teacher network to regularize feature learning and quantifies data-dependent aleatoric uncertainty by adapting the network to predict intrinsic input variances. Experiments on the LGE MRI left atrium dataset and Abdomen CT-1K liver datasets demonstrate that our technique enhances the accuracy and robustness of sparse image-driven SSM.

Decomposed Prototype Learning for Few-Shot Scene Graph Generation

Today's scene graph generation (SGG) models typically require abundant manual annotations to learn new predicate types. Therefore, it is difficult to apply them to real-world applications with massive uncommon predicate categories whose annotations are hard to collect. In this paper, we focus on Few-Shot SGG (FSSGG) , which encourages SGG models to be able to quickly transfer previous knowledge and recognize unseen predicates well with only a few examples. However, current methods for FSSGG are hindered by the high intra-class variance of predicate categories in SGG: On one hand, each predicate category commonly has multiple semantic meanings under different contexts. On the other hand, the visual appearance of relation triplets with the same predicate differs greatly under different subject-object compositions. Such great variance of inputs makes it hard to learn generalizable representation for each predicate category with current few-shot learning (FSL) methods. However, we found that this intra-class variance of predicates is highly related to the composed subjects and objects. To model the intra-class variance of predicates with subject-object context, we propose a novel Decomposed Prototype Learning (DPL) model for FSSGG. Specifically, we first construct a decomposable prototype space to capture diverse semantics and visual patterns of subjects and objects for predicates by decomposing them into multiple prototypes. Afterwards, we integrate these prototypes with different weights to generate query-adaptive predicate representation with more reliable semantics for each query sample. We conduct extensive experiments and compare with various baseline methods to show the effectiveness of our method.

Going beyond still images to improve input variance resilience in multi-stream vision understanding models

Traditionally, vision models have predominantly relied on spatial features extracted from static images, deviating from the continuous stream of spatiotemporal features processed by the brain in natural vision. While numerous video-understanding models have emerged, incorporating videos into image-understanding models with spatiotemporal features has been limited. Drawing inspiration from natural vision, which exhibits remarkable resilience to input changes, our research focuses on the development of a brain-inspired model for vision understanding trained with videos. Our findings demonstrate that models that train on videos instead of still images and include temporal features become more resilient to various alternations on input media.

Extensive data engineering to the rescue: building a multi-species katydid detector from unbalanced, atypical training datasets.

Passive acoustic monitoring (PAM) is a powerful tool for studying ecosystems. However, its effective application in tropical environments, particularly for insects, poses distinct challenges. Neotropical katydids produce complex species-specific calls, spanning mere milliseconds to seconds and spread across broad audible and ultrasonic frequencies. However, subtle differences in inter-pulse intervals or central frequencies are often the only discriminatory traits. These extremities, coupled with low source levels and susceptibility to masking by ambient noise, challenge species identification in PAM recordings. This study aimed to develop a deep learning-based solution to automate the recognition of 31 katydid species of interest in a biodiverse Panamanian forest with over 80 katydid species. Besides the innate challenges, our efforts were also encumbered by a limited and imbalanced initial training dataset comprising domain-mismatched recordings. To overcome these, we applied rigorous data engineering, improving input variance through controlled playback re-recordings and by employing physics-based data augmentation techniques, and tuning signal-processing, model and training parameters to produce a custom well-fit solution. Methods developed here are incorporated into Koogu, an open-source Python-based toolbox for developing deep learning-based bioacoustic analysis solutions. The parametric implementations offer a valuable resource, enhancing the capabilities of PAM for studying insects in tropical ecosystems. This article is part of the theme issue 'Towards a toolkit for global insect biodiversity monitoring'.

Bayesian inference is facilitated by modular neural networks with different time scales.

Various animals, including humans, have been suggested to perform Bayesian inferences to handle noisy, time-varying external information. In performing Bayesian inference by the brain, the prior distribution must be acquired and represented by sampling noisy external inputs. However, the mechanism by which neural activities represent such distributions has not yet been elucidated. Our findings reveal that networks with modular structures, composed of fast and slow modules, are adept at representing this prior distribution, enabling more accurate Bayesian inferences. Specifically, the modular network that consists of a main module connected with input and output layers and a sub-module with slower neural activity connected only with the main module outperformed networks with uniform time scales. Prior information was represented specifically by the slow sub-module, which could integrate observed signals over an appropriate period and represent input means and variances. Accordingly, the neural network could effectively predict the time-varying inputs. Furthermore, by training the time scales of neurons starting from networks with uniform time scales and without modular structure, the above slow-fast modular network structure and the division of roles in which prior knowledge is selectively represented in the slow sub-modules spontaneously emerged. These results explain how the prior distribution for Bayesian inference is represented in the brain, provide insight into the relevance of modular structure with time scale hierarchy to information processing, and elucidate the significance of brain areas with slower time scales.

An empirical study on the effect of academic engagement on academic achievement of engineering students in colleges and universities

Abstract Based on the linear regression algorithm, this paper uses the basic linear regression algorithm and the kernel linear regression algorithm to construct the relationship model between academic input and academic achievement based on linear regression and to determine the independent variables, dependent variables and control variables in the model. The engineering students of East China Normal University are selected as the research object, the research instrument is determined, the research index is determined according to the survey, and the research hypothesis is proposed, and then the research and analysis design of the impact of academic input on academic achievement is completed. Using reliability and validity to test the validity of the questionnaire, the initial data of the study can be obtained by distributing the questionnaire, and correlation analysis and regression analysis are used to carry out an empirical study of academic input and academic achievement based on linear regression. The results show that the correlation coefficient between academic input and academic achievement of master’s degree students is R=0.892 and R2=0.792, indicating that the predictive variance of academic input on academic achievement is 89.2%. The effects of the dimensions under the academic input on the social adjustment dimension under the academic achievement were, in descending order, as follows: active cooperation > teacher-student interaction > educational enrichment > academic challenge > campus support. This study comprehensively analyzes the impact of academic engagement on academic achievement, and colleges and universities should change the test of students’ ability to memorize knowledge and skills to the test of students’ ability to analyze and solve problems by applying what they have learned.

An unscented Kalman filter under unknown input without direct feedthrough for joint input and system identification of structural systems

An unscented Kalman filter under unknown input without direct feedthrough for joint input and system identification of structural systems

The Difference Method Approach for Sampling Order Constrained Parameters: An Improved Implementation and Important Limitations.

A health economic model may include a set of related inputs whose true values are uncertain, but that can be assumed to follow a logical order. Various approaches are available for performing probabilistic sensitivity analysis while preserving the order constraint, one such approach is known as the difference method. The difference method approach appears to have many of the required properties, has been endorsed by good practice guidelines, and is likely to prove a popular approach. However, the proposed implementation of the difference method approach is cumbersome, requiring numerical estimation, which might present a barrier to its adoption. Furthermore, it is unclear whether the method can always be applied to three or more model inputs and whether it is unbiased across all possible input values. This study has investigated these three issues for ordered inputs bounded between 0 and 1. An analytic solution is given that allows for more straightforward and compact implementation. The difference method approach cannot always be applied to a set of three or more model inputs, and this depends on the relative size of the variances of the logit-transformed Beta distributions fitted to each variable. The approach can also produce samples with biased means and variances under certain combinations of input means and variances. It is recommended that the difference method approach be used where appropriate; however, an understanding of its limitations is necessary to identify such cases.

Multitasking via baseline control in recurrent neural networks

Changes in behavioral state, such as arousal and movements, strongly affect neural activity in sensory areas, and can be modeled as long-range projections regulating the mean and variance of baseline input currents. What are the computational benefits of these baseline modulations? We investigate this question within a brain-inspired framework for reservoir computing, where we vary the quenched baseline inputs to a recurrent neural network with random couplings. We found that baseline modulations control the dynamical phase of the reservoir network, unlocking a vast repertoire of network phases. We uncovered a number of bistable phases exhibiting the simultaneous coexistence of fixed points and chaos, of two fixed points, and of weak and strong chaos. We identified several phenomena, including noise-driven enhancement of chaos and ergodicity breaking; neural hysteresis, whereby transitions across a phase boundary retain the memory of the preceding phase. In each bistable phase, the reservoir performs a different binary decision-making task. Fast switching between different tasks can be controlled by adjusting the baseline input mean and variance. Moreover, we found that the reservoir network achieves optimal memory performance at any first-order phase boundary. In summary, baseline control enables multitasking without any optimization of the network couplings, opening directions for brain-inspired artificial intelligence and providing an interpretation for the ubiquitously observed behavioral modulations of cortical activity.

A Non-Linear Body Machine Interface for Controlling Assistive Robotic Arms.

Body machine interfaces (BoMIs) enable individuals with paralysis to achieve a greater measure of independence in daily activities by assisting the control of devices such as robotic manipulators. The first BoMIs relied on Principal Component Analysis (PCA) to extract a lower dimensional control space from information in voluntary movement signals. Despite its widespread use, PCA might not be suited for controlling devices with a large number of degrees of freedom, as because of PCs' orthonormality the variance explained by successive components drops sharply after the first. Here, we propose an alternative BoMI based on non-linear autoencoder (AE) networks that mapped arm kinematic signals into joint angles of a 4D virtual robotic manipulator. First, we performed a validation procedure that aimed at selecting an AE structure that would allow to distribute the input variance uniformly across the dimensions of the control space. Then, we assessed the users' proficiency practicing a 3D reaching task by operating the robot with the validated AE. All participants managed to acquire an adequate level of skill when operating the 4D robot. Moreover, they retained the performance across two non-consecutive days of training. While providing users with a fully continuous control of the robot, the entirely unsupervised nature of our approach makes it ideal for applications in a clinical context since it can be tailored to each user's residual movements. We consider these findings as supporting a future implementation of our interface as an assistive tool for people with motor impairments.

Hepatic follistatin increases basal metabolic rate and attenuates diet-induced obesity during hepatic insulin resistance

Body weight change and obesity follow the variance of excess energy input balanced against tightly controlled EE (energy expenditure). Since insulin resistance can reduce energy storage, we investigated whether genetic disruption of hepatic insulin signaling reduced adipose mass with increased EE. Insulin signaling was disrupted by genetic inactivation of Irs1 (Insulin receptor substrate 1) and Irs2 in hepatocytes of LDKO mice (Irs1L/L·Irs2L/L·CreAlb), creating a state of complete hepatic insulin resistance. We inactivated FoxO1 or the FoxO1-regulated hepatokine Fst (Follistatin) in the liver of LDKO mice by intercrossing LDKO mice with FoxO1L/L or FstL/L mice. We used DEXA (dual-energy X-ray absorptiometry) to assess total lean mass, fat mass and fat percentage, and metabolic cages to measure EE (energy expenditure) and estimate basal metabolic rate (BMR). High-fat diet was used to induce obesity. Hepatic disruption of Irs1 and Irs2 (LDKO mice) attenuated HFD (high-fat diet)-induced obesity and increased whole-body EE in a FoxO1-dependent manner. Hepatic disruption of the FoxO1-regulated hepatokine Fst normalized EE in LDKO mice and restored adipose mass during HFD consumption; moreover, hepatic Fst disruption alone increased fat mass accumulation, whereas hepatic overexpression of Fst reduced HFD-induced obesity. Excess circulating Fst in overexpressing mice neutralized Mstn (Myostatin), activating mTORC1-promoted pathways of nutrient uptake and EE in skeletal muscle. Similar to Fst overexpression, direct activation of muscle mTORC1 also reduced adipose mass. Thus, complete hepatic insulin resistance in LDKO mice fed a HFD revealed Fst-mediated communication between the liver and muscle, which might go unnoticed during ordinary hepatic insulin resistance as a mechanism to increase muscle EE and constrain obesity.

How subcultures emerge.

Sympatric speciation is typically presented as a rare phenomenon, but urban subcultures frequently emerge even in the absence of geographic isolation. Is there perhaps something that culture has but biological inheritance does not that would account for this difference? We present a novel model that combines assortative interaction and multidimensional inheritance. Our computer simulations show that assortment alone can lead to the formation of cohesive clusters of individuals with low within-group and large between-group variability even in the absence of a spatial separation or disruptive natural selection. All it takes is a proportionality between the variance of inputs (cultural 'parents') and outputs (cultural 'offspring'). We argue that variability-dependent inheritance cannot be easily accomplished by genes alone, but it may be the norm, not the exception, in the transmission of culture between humans. This model explains the frequent emergence of subcultures and behavioural clustering in our species and possibly also other cultural animals.

Finite Impulse Response Errors-in-Variables System Identification Utilizing Approximated Likelihood and Gaussian Mixture Models

In this paper a Maximum likelihood estimation algorithm for Finite Impulse Response Errors-in-Variables systems is developed. We consider that the noise-free input signal is Gaussian-mixture distributed. We propose an Expectation-Maximization-based algorithm to estimate the system model parameters, the input and output noise variances, and the Gaussian mixture noise-free input parameters. The benefits of our proposal are illustrated via numerical simulations.

Development and assessment of a model predictive controller enabling anticipatory control strategies for a heat-pipe system

To support the reliable and resilient operation of modular reactors and microreactors, anticipatory control strategies have been proposed for achieving faster-than-real-time predictions and decision-making capabilities in anticipation of potential anomalies, including setpoint changes and cyber incidents. This work presents how anticipatory control strategies can be implemented via model predictive control (MPC) for a single heat pipe's temperature. Considering the uncertainty in developing and applying MPC, this work evaluates MPC performance given three different model forms: a linear response surface model, an artificial neural network (ANN), and an autoregressive model with exogenous input (ARX). This work also evaluates the impacts of different input biases and variance on MPC performance in order to account for potential sensor reading variations due to cyber incidents. We observe that the ANN and ARX result in more fluctuated control actions compared to the MPC with linear response surface model. However, when the cyber incidents are of large magnitudes, the linear response surface model produces smaller feasible regions than the ANN and ARX models under identical constraints.

An emergent temporal basis set robustly supports cerebellar time-series learning.

The cerebellum is considered a "learning machine" essential for time interval estimation underlying motor coordination and other behaviors. Theoretical work has proposed that the cerebellum's input recipient structure, the granule cell layer (GCL), performs pattern separation of inputs that facilitates learning in Purkinje cells (P-cells). However, the relationship between input reformatting and learning has remained debated, with roles emphasized for pattern separation features from sparsification to decorrelation. We took a novel approach by training a minimalist model of the cerebellar cortex to learn complex time-series data from time-varying inputs, typical during movements. The model robustly produced temporal basis sets from these inputs, and the resultant GCL output supported better learning of temporally complex target functions than mossy fibers alone. Learning was optimized at intermediate threshold levels, supporting relatively dense granule cell activity, yet the key statistical features in GCL population activity that drove learning differed from those seen previously for classification tasks. These findings advance testable hypotheses for mechanisms of temporal basis set formation and predict that moderately dense population activity optimizes learning.NEW & NOTEWORTHY During movement, mossy fiber inputs to the cerebellum relay time-varying information with strong intrinsic relationships to ongoing movement. Are such mossy fibers signals sufficient to support Purkinje signals and learning? In a model, we show how the GCL greatly improves Purkinje learning of complex, temporally dynamic signals relative to mossy fibers alone. Learning-optimized GCL population activity was moderately dense, which retained intrinsic input variance while also performing pattern separation.

Accuracy and uncertainty of geostatistical models versus machine learning for digital mapping of soil calcium and potassium.

Accuracy and uncertainty of models used for digital soil mapping are important for assessing confidence of predictions and reliable land use planning and management. In this study, two approaches of geostatistical (spatial) and machine learning (ML) models were evaluated forpredictivemapping of soil calcium (Ca) and potassium (K). Two spatial models including empirical Bayesian kriging (EBK) and sequential Gaussian simulation (SGS) were compared with machine learning models: Cubist, random forest (RF) and support vector machine (SVM) in terms of their accuracy and uncertainty for mapping soil Ca and K. The study area is in Nowley, New South Wales, Australia, with an area of 2083ha and a variety of soil types and farming systems. For the models training process, 240 soil samples data and for validation 102 independent samples data were used. For accuracy assessment R2, root mean square error (RMSE), concordance and bias and for uncertainty assessment confidence limits were investigated. Also, in order to compare the outcomes for the two soil properties with different measurement units, mean absolute percentage error (MAPE) and relative uncertainty (RU) as accuracy and uncertainty measures, respectively, were evaluated. Results showed that for K map SGS had the highest R2 (0.74) and lowest RMSE (1.96), followed by EBK with R2 = 0.72 and RMSE = 2.02. For Ca map, EBK model showed the highest accuracy (R2 = 0.46; RMSE = 3.21), followed by SVM and SGS with comparable accuracies. Comparing the two soil properties, Ca map showed higher MAPE and RU, compared to K map. The lowest MAPE was obtained for EBK model (for K = 39) and SGS model (for K = 44). Also, the lowest RU values were found for EBK and SGS models. Among the ML models, SVM showed lower sensitivity to higher variance in data input. In general, the spatial models outperformed the ML models with regard to both accuracy and uncertainty. An additional conclusion is that considering the data variance in the two soil properties, geostatistical models with lower RU and MAPE were relatively less susceptible to data variance, compared to ML models.

Efficient Stochastic Galerkin Spectral Methods for Optimal Control Problems Constrained by Fractional PDEs with Uncertain Inputs

This paper is devoted to designing fast solvers and efficient preconditioners for the optimal control problems (OCPs) constrained by stochastic fractional elliptic equations. We first prove the existence and uniqueness of the stochastic optimal control solution and then derive the stochastic optimality system. For the numerical approximation, we use the stochastic Galerkin spectral methods, which apply the stochastic Galerkin method for the discretization of random variables and employ the spectral-Galerkin approach for the approximation of spatial variables. To solve the large coupled saddle-point system resulted from discretization, we adopt the most commonly used MINRES method and a more effective PPCG method in low-rank matrix iteration format. Specially, we develop the efficient preconditioners based on the matrix decomposition method and the virtual variable method. We also study the eigenvalue distribution of the corresponding preconditioned matrix. Besides, for the approximation of the discretized state equation, we use the mean-based approximation and the Ullmann approximation to handle different values of the variance of the random inputs. Finally, we present numerical experiments to demonstrate the effectiveness of our solvers and preconditioners.

Training much deeper spiking neural networks with a small number of time-steps

Spiking Neural Network (SNN) is a promising energy-efficient neural architecture when implemented on neuromorphic hardware. The Artificial Neural Network (ANN) to SNN conversion method, which is the most effective SNN training method, has successfully converted moderately deep ANNs to SNNs with satisfactory performance. However, this method requires a large number of time-steps, which hurts the energy efficiency of SNNs. How to effectively covert a very deep ANN (e.g., more than 100 layers) to an SNN with a small number of time-steps remains a difficult task. To tackle this challenge, this paper makes the first attempt to propose a novel error analysis framework that takes both the "quantization error" and the "deviation error" into account, which comes from the discretization of SNN dynamicsthe neuron's coding scheme and the inconstant input currents at intermediate layers, respectively. Particularly, our theories reveal that the "deviation error" depends on both the spike threshold and the input variance. Based on our theoretical analysis, we further propose the Threshold Tuning and Residual Block Restructuring (TTRBR) method that can convert very deep ANNs (>100 layers) to SNNs with negligible accuracy degradation while requiring only a small number of time-steps. With very deep networks, our TTRBR method achieves state-of-the-art (SOTA) performance on the CIFAR-10, CIFAR-100, and ImageNet classification tasks.

Analysis of input set characteristics and variances on k-fold cross validation for a Recurrent Neural Network model on waste disposal rate estimation

The use of machine learning techniques in waste management studies is increasingly popular. Recent literature suggests k-fold cross validation may reduce input dataset partition uncertainties and minimize overfitting issues. The objectives are to quantify the benefits of k-fold cross validation for municipal waste disposal prediction and to identify the relationship of testing dataset variance on predictive neural network model performance. It is hypothesized that the dataset characteristics and variances may dictate the necessity of k-fold cross validation on neural network waste model construction. Seven RNN-LSTM predictive models were developed using historical landfill waste records and climatic and socio-economic data. The performance of all trials was acceptable in the training and validation stages, with MAPE all less than 10%. In this study, the 7-fold cross validation reduced the bias in selection of testing sets as it helps to reduce MAPE by up to 44.57%, MSE by up to 54.15%, and increased R value by up to 8.33%. Correlation analysis suggests that fewer outliers and less variance of the testing dataset correlated well with lower modeling error. The length of the continuous high waste season and length of total high waste period appear not important to the model performance. The result suggests that k-fold cross validation should be applied to testing datasets with higher variances. The use of MSE as an evaluation index is recommended.