arXiv subjects

Bert Debusschere

Publications and source records attributed to Bert Debusschere.

4 recordsLinked to original sources

arXiv2026

Materials Behavior as Mechanism Ensembles: A Probabilistic Framework for Emergent Behaviors

Materials behavior is often treated as a deterministic mapping from structure to properties, yet many important phenomena emerge from the conditional activation of multiple mechanisms across scales. This is especially evident in fatigue of metals, where crack growth is typically modeled as monotonic and irreversible process, despite evidence that local microstructure, loading history, and competing unit processes can shift the balance among propagation, arrest, and self-healing. Here we present a probabilistic framework that describes materials behavior as an ensemble of constituent mechanisms whose activation, interaction, and evolution determine emergent outcomes. The framework connects mechanism activation, state evolution, and macroscopic observables in a probabilistic way. In the case of fatigue crack propagation, it reframes damage tolerance as an inference problem over mechanism competition and provides a basis for integrating multiscale simulation, multimodal characterization, and machine learning. The same logic extends to other physical and chemical systems suggesting a portable framework for any system in which emergent behavior reflects mechanism competition under changing conditions. The broader ambition of this perspective review is a shift from correlating structure and performance after the fact to identifying, in advance, the conditions that make desired emergent behavior probable.

cond-mat.mtrl-sci↗

arXiv2025

Scalable extensions to given-data Sobol' index estimators

Given-data methods for variance-based sensitivity analysis have significantly advanced the feasibility of Sobol' index computation for computationally expensive models and models with many inputs. However, the limitations of existing methods still preclude their application to models with an extremely large number of inputs. In this work, we present practical and theoretical extensions to the existing given-data Sobol' index method, which allow variance-based sensitivity analysis to be efficiently performed on large models such as neural networks, which have $>10^4$ inputs. For models of this size, holding all input-output evaluations simultaneously in memory---as required by existing methods---can quickly become impractical. Our extensions include a general definition of the given-data Sobol' index estimator with arbitrary partition, a streaming algorithm to process input-output samples in batches, and an asymptotic analysis of the new estimator that motivates a practical screening heuristic for small indices. We show that the equiprobable partition employed in existing given-data methods can introduce significant bias into Sobol' index estimates even at large sample sizes and provide numerical analyses that demonstrate why this can occur. We also show that the streaming algorithm can achieve comparable accuracy and runtime while substantially reducing memory requirements, enabling sensitivity analysis of models with much larger input dimension. We demonstrate our novel developments on two application problems in neural network modeling.

stat.ML↗

arXiv2025

Improving the quasi-biennial oscillation via a surrogate-accelerated multi-objective optimization

Simulating the QBO remains a formidable challenge partly due to uncertainties in representing convectively generated gravity waves. We develop an end-to-end uncertainty quantification workflow that calibrates these gravity wave processes in E3SM to yield a more realistic QBO. Central to our approach is a domain knowledge-informed, compressed representation of high-dimensional spatio-temporal wind fields. By employing a parsimonious statistical model that learns the fundamental frequency of the underlying stochastic process from complex observations, we extract a concise set of interpretable and physically meaningful quantities of interest capturing key attributes, such as oscillation amplitude and period. Building on this, we train a probabilistic surrogate model. Leveraging the Karhunen-Loeve decomposition, our surrogate efficiently represents these characteristics as a set of orthogonal features, thereby capturing the cross-correlations among multiple physics quantities evaluated at different stratospheric pressure levels, and enabling rapid surrogate-based inference at a fraction of the computational cost of inference reliant only on full-scale simulations. Finally, we analyze the inverse problem using a multi-objective approach. Our study reveals a tension between amplitude and period that constrains the QBO representation, precluding a single optimal solution that simultaneously satisfies both objectives. To navigate this challenge, we quantify the bi-criteria trade-off and generate a representative set of Pareto optimal physics parameter values that balance the conflicting objectives. This integrated workflow not only improves the fidelity of QBO simulations but also advances toward a practical framework for tuning modes of variability and quasi-periodic phenomena, offering a versatile template for uncertainty quantification in complex geophysical models.

physics.ao-ph↗

arXiv2022

An out-of-distribution discriminator based on Bayesian neural network epistemic uncertainty

Neural networks have revolutionized the field of machine learning with increased predictive capability. In addition to improving the predictions of neural networks, there is a simultaneous demand for reliable uncertainty quantification on estimates made by machine learning methods such as neural networks. Bayesian neural networks (BNNs) are an important type of neural network with built-in capability for quantifying uncertainty. This paper discusses aleatoric and epistemic uncertainty in BNNs and how they can be calculated. With an example dataset of images where the goal is to identify the amplitude of an event in the image, it is shown that epistemic uncertainty tends to be lower in images which are well-represented in the training dataset and tends to be high in images which are not well-represented. An algorithm for out-of-distribution (OoD) detection with BNN epistemic uncertainty is introduced along with various experiments demonstrating factors influencing the OoD detection capability in a BNN. The OoD detection capability with epistemic uncertainty is shown to be comparable to the OoD detection in the discriminator network of a generative adversarial network (GAN) with comparable network architecture.

cs.LG↗