Explore the vast range of titles available.

The whole sentence representation reasoning process simultaneously comprises a sentence representation module and a semantic reasoning module. This paper combines the multi-layer semantic representation network with the deep fusion matching network to solve the limitations of only considering a sentence representation module or a reasoning model. It proposes a joint optimization method based on multi-layer semantics called the Semantic Fusion Deep Matching Network (SCF-DMN) to explore the influence of sentence representation and reasoning models on reasoning performance. Experiments on text entailment recognition tasks show that the joint optimization representation reasoning method performs better than the existing methods. The sentence representation optimization module and the improved optimization reasoning model can promote reasoning performance when used individually. However, the optimization of the reasoning model has a more significant impact on the final reasoning results. Furthermore, after comparing each module’s performance, there is a mutual constraint between the sentence representation module and the reasoning model. This condition restricts overall performance, resulting in no linear superposition of reasoning performance. Overall, by comparing the proposed methods with other existed methods that are tested using the same database, the proposed method solves the lack of in-depth interactive information and interpretability in the model design which would be inspirational for future improving and studying of natural language reasoning.

We show that several machine learning estimators, including square-root least absolute shrinkage and selection and regularized logistic regression, can be represented as solutions to distributionally robust optimization problems. The associated uncertainty regions are based on suitably defined Wasserstein distances. Hence, our representations allow us to view regularization as a result of introducing an artificial adversary that perturbs the empirical distribution to account for out-of-sample effects in loss estimation. In addition, we introduce RWPI (robust Wasserstein profile inference), a novel inference methodology which extends the use of methods inspired by empirical likelihood to the setting of optimal transport costs (of which Wasserstein distances are a particular case). We use RWPI to show how to optimally select the size of uncertainty regions, and as a consequence we are able to choose regularization parameters for these machine learning estimators without the use of cross validation. Numerical experiments are also given to validate our theoretical findings.

Machine learned force fields typically require manual construction of training sets consisting of thousands of first principles calculations, which can result in low training efficiency and unpredictable errors when applied to structures not represented in the training set of the model. This severely limits the practical application of these models in systems with dynamics governed by important rare events, such as chemical reactions and diffusion. We present an adaptive Bayesian inference method for automating the training of interpretable, low-dimensional, and multi-element interatomic force fields using structures drawn on the fly from molecular dynamics simulations. Within an active learning framework, the internal uncertainty of a Gaussian process regression model is used to decide whether to accept the model prediction or to perform a first principles calculation to augment the training set of the model. The method is applied to a range of single- and multi-element systems and shown to achieve a favorable balance of accuracy and computational efficiency, while requiring a minimal amount of ab initio training data. We provide a fully open-source implementation of our method, as well as a procedure to map trained models to computationally efficient tabulated force fields.

Understanding the complex biosynthetic pathways of glycosylation is crucial for the expanding field of glycosciences. Computer-aided glycosylation analysis has greatly benefited in recent years from the development of tools found in web-based portals and open-source libraries. However, the in silico analysis of cellular glycosylation kinetics is underrepresented in current glycoscience-related tools and databases. This could be partly attributed to the limited accessibility of kinetic models developed using proprietary software and the difficulty in reliably parameterising such models. This work aims to address these challenges by proposing GlyCompute, an open-source framework demonstrating a novel, streamlined approach for the assembly, simulation, and parameterisation of kinetic models of protein N-linked glycosylation. Specifically, given one or more sets of experimentally observed N-glycan structures and their relative abundances, minimum representations of a glycosylation reaction network are generated. The topology of the resulting networks is then used to automatically assemble the material balances and kinetic mechanisms underpinning the mathematical model. To match the experimentally observed relative abundances, a sequential parameter estimation strategy using Bayesian inference is proposed, with stages determined automatically based on the underlying network topology. The proposed framework was tested on a case study involving the simultaneous fitting of the kinetic model to two protein N-linked glycoprofiles produced by the same CHO cell culture, showing good agreement with experimental observations. We envision that GlyCompute could help glycoscientists gain quantitative insights into the effect of enzyme kinetics and their perturbations on experimentally observed glycoprofiles in biomanufacturing and clinical settings. Graphical Abstract

Nonlinear Tikhonov regularization within a Bayesian framework is incorporated into a computer program called pyReSpect, which infers the continuous and discrete relaxation spectra from oscillatory shear experiments. It uses Bayesian inference to provide uncertainty estimates for the continuous spectrum h ( τ ) by propagating the uncertainty in the regularization parameter λ . The new algorithm is about 6–9 times faster than an older version of the program (ReSpect) in which the optimal λ was determined by the L-curve method. About half of the speedup arises from the Bayesian formulation by restricting the window of λ explored. The other half arises from the nonlinear formulation for which the spectrum is a weak function of λ , allowing us to use a coarse mesh for λ . The program is tested and validated on three examples: a synthetic spectrum, a H-polymer, and an elastomer with a nonzero terminal plateau.

We develop several inference methods to estimate the position of dislocations from images generated using dark-field X-ray microscopy (DFXM)—achieving superresolution accuracy and principled uncertainty quantification. Using the framework of Bayesian inference, we incorporate models of the DFXM contrast mechanism and detector measurement noise, along with initial position estimates, into a statistical model coupling DFXM images with the dislocation position of interest. We motivate several position estimation and uncertainty quantification algorithms based on this model. We then demonstrate the accuracy of our primary estimation algorithm on synthetic realistic DFXM images of edge dislocations in single-crystal aluminum. We conclude with a discussion of our methods’ impact on future dislocation studies and possible future research avenues.

Composite plates are susceptible to various damages in complex conditions and working environments, which may reduce the reliability of the structure and threaten equipment and personal safety. Thus, the implementation of a robust online Structural health monitoring (SHM) system for these composite structures becomes imperative. To enhance reliability and safety, we introduce a robust online SHM system anchored by our newly developed damage detection Bayesian neural network (DD-BNN). The main contribution of this study lies in the DD-BNN to perform precise and reliable damage detection and localization in composite plates using only one actuator-receiver pair without any signal/feature pre-processing and human intervention. The proposed DD-BNN model innovatively combines probabilistic modeling with deep learning to address uncertainty in Lamb wave-based damage detection and model performance for composite plates, featuring a specialized probabilistic layer trained through Bayesian inference to efficiently encapsulate and manage uncertainty in model weights and activation. Notably, our method significantly simplifies the SHM system design and manual operation requirements. In addition, this approach not only reduces overfitting but also enhances robustness to noise, as confirmed by experiments on perturbation analysis of Gaussian and Poisson noise.

Machine-learned interatomic potentials enable realistic finite temperature calculations of complex materials properties with first-principles accuracy. It is not yet clear, however, how accurately they describe anharmonic properties, which are crucial for predicting the lattice thermal conductivity and phase transitions in solids and, thus, shape their technological applications. Here we employ a recently developed on-the-fly learning technique based on molecular dynamics and Bayesian inference in order to generate an interatomic potential capable to describe the thermodynamic properties of zirconia, an important transition metal oxide. This machine-learned potential accurately captures the temperature-induced phase transitions below the melting point. We further showcase the predictive power of the potential by calculating the heat transport on the basis of Green–Kubo theory, which allows to account for anharmonic effects to all orders. This study indicates that machine-learned potentials trained on the fly offer a routine solution for accurate and efficient simulations of the thermodynamic properties of a vast class of anharmonic materials.

This study investigates the causal effects of changing the atomic concentration of a single element within high-entropy alloys (HEAs) on their yield strength. We propose a causal inference model for HEAs, named HEACM, and estimate the effects using the double/debiased machine learning framework. Four machine learning methods, including the generalized linear model (GLM), random forests (RF), support vector machine (SVM), and extreme gradient boosting (XGB), are used as estimators within HEACM. We assess the effects of changes in valence electron concentration, mixing enthalpy, and mixing entropy on yield strength resulting from these atomic concentration changes. Experimental results indicate that a positive unit change in valence electron concentration and mixing enthalpy decreases yield strength by averages of 252.3 MPa and 29.8 MPa, respectively, while a positive unit change in mixing entropy results in an average increase in yield strength of 217.4 MPa. The XGB method provides the smallest 95% confidence interval size, suggesting it is the most reliable estimator.

Collagen Type I protein-based scaffolds are most commonly used because of their better healing properties. Production of type I collagen nanofibers is a challenge because of the high molecular weight of collagen obtained using the electrospinning method. For this reason, biodegradable polyvinyl (alchool) (PVA) was employed as an electrospun collagen type I source for bovine tendon. The aim of this study was to model the average diameter (D a ) of PVA/collagen type-I nanofibers using the input parameters PVA ratio (P), collagen type I ratio (C), PVA mixing ratio/collagen type I mixing ratio (M r ), applied voltage (V), and feed rate (F r ) parameters. The specified parameters were designed to produce nanofibers with various diameters. Scanning electron microscopy (SEM) with image software was used to calculate the D a of the nanofibers. The mean squared error (MSE) and coefficient of determination (R 2 ) were evaluated for the observed and predicted nanofiber diameters. The findings show that increasing the ratio of PVA from 7 wt% to 9 wt% rise ~ 7% the average diameter. The average nanofiber diameter of PVA/collagen type I was varied according to the mixture, quantity of collagen, and electrospinning parameters. Furthermore, the D a of the nanofibers increased with increasing collagen ratio and mixing ratio. In adverse cases, PVA (8 wt%) with Collagen Type I (4 wt%) having a ratio of 2:1 exhibited lower D a with 76.2 nm. The observed results indicated that the designed adaptive neuro-fuzzy inference system model was effective for evaluating the relevant parameters and predicting the D a of PVA/collagen type I.