#uncertainty quantification
45 resultsCalibrated uncertainty for wide-angle crustal models: how firmly is the Dharwar Craton Moho actually constrained?
Deepak Kumar, Laxmidhar Behera, Wojciech Czuba
The paper assesses how well the Moho depth beneath the Dharwar Craton is constrained by applying calibrated uncertainty analysis and bootstrap validation to a large wide-angle seis…
Reliability-calibrated deep residual full-waveform inversion using geometry-invariant physics encoding: synthetic validation and zero-shot Marmousi-2 testing
Deepak Kumar, Jayant Nath Tripathi, Laxmidhar Behera
The paper presents a deep learning pipeline that uses physics‑derived inputs to correct full‑waveform inversion results and provides calibrated, pixel‑wise uncertainty estimates th…
Uncertainty quantification for trustworthy deep learning: Methods and measures
H. Martin Gillis, Thomas Trappenberg
The paper surveys methods for quantifying uncertainty in deep neural networks, focusing on ensemble-based and approximate Bayesian approaches and how their outputs are measured.
Cost-Sensitive Conformal Prediction and Human-in-the-Loop Abstention for Imbalanced High-Stakes Decision Support: A Multi-Domain Benchmark
Manpreet Singh, Akshatha Srikantha, Shyamal Lakhanpal
The paper evaluates how conformal prediction methods can be made cost‑sensitive and reliable for rare, high‑cost classes in imbalanced, high‑stakes decision tasks, and shows that M…
From Keypoints to Predictive Distributions: Post-Hoc Uncertainty for YOLO-Pose Models
Alexej Klushyn, Juan Rivero Sesma, Florian Seligmann +3
The paper adds a lightweight post‑hoc module to trained YOLO‑Pose models that predicts calibrated predictive distributions for each keypoint, enabling uncertainty‑aware ranking and…
Decoding the Micromagnetic Hamiltonian from Magnetic Fingerprints
Bradley J. Fugetta, Anqi Liu, Kai Liu +2
The paper presents deep convolutional neural networks that infer the full micromagnetic Hamiltonian of a material directly from magnetic fingerprint data obtained via First‑Order R…
Runtime Uncertainty Monitoring for LLM-Based Multi-Agent Systems Using Bayesian Networks
Bart Custers, Koorosh Aslansefat
The paper presents a framework for monitoring runtime uncertainty in large‑language‑model based multi‑agent systems by converting token‑level log‑probabilities into calibrated conf…
MetaKoopman: Bayesian Meta-Learning of Koopman Operators for Modeling Structured Dynamics under Distribution Shifts
Mahmoud Selim, Sriharsha Bhat, Karl H. Johansson
MetaKoopman is a Bayesian meta‑learning framework that learns a prior over Koopman operators to model nonlinear dynamics with linear latent representations, providing closed‑form u…
On the optimality of dimension truncation error rates for a class of parametric partial differential equations
Philipp A. Guth, Vesa Kaarnioja
The paper analyzes the error introduced when infinite-dimensional random field inputs in parametric PDEs are truncated to finite dimensions, and proves that the known dimension‑tru…
On Hyperbolic Stochastic Galerkin Projections of Shallow Water Linearised Moment Equations
Safiere Kuijpers, Julian Koellermeier
The paper develops an intrusive stochastic Galerkin method for one‑dimensional shallow water linearised moment equations, introduces a regularisation to preserve hyperbolicity, and…
Finite-Sample Conformal Coverage Recovery via Fusion under Degraded Local Guarantees in Occupancy Map Estimation
Ritvik Mahajan, Aneesh Raghavan, Karl Henrik Johansson
The paper proposes a distributed fusion method that combines locally calibrated conformal prediction maps from multiple robots to achieve a user‑specified coverage guarantee in occ…
Evaluating Epistemic Uncertainty: Beyond OOD Detection and Active Learning
Jakub Paplhám, Willem Waegeman, Eyke Hüllermeier +1
The paper proposes a decision‑theoretic framework for evaluating epistemic uncertainty by measuring its ability to identify reducible error (regret) in selective prediction, and sh…
Statistical Inference for Scenario-Based Dynamic Optimization under Uncertainty
Aurya Javeed, Johannes Milz
The paper develops statistical methods for evaluating and constructing confidence intervals for the optimal value of finite-horizon open-loop dynamic optimization problems when par…
Subjective Risk Decomposition: A New View for Uncertainty Quantification
Raghad Alamri, Michele Caprio, Gavin Brown
The paper introduces a framework that derives epistemic and aleatoric uncertainty measures by decomposing a subjective risk defined via a strictly proper loss, unifying many existi…
Contrastive Conformal Sets
Yahya Alkhatib, Wee Peng Tay
The paper introduces a method that combines contrastive learning with conformal prediction to create learnable geometric sets that guarantee a user‑specified coverage of positive s…
Gradient-enhanced spline dimensional decomposition for uncertainty quantification with limited training samples
Eunho Heo, Dongjin Lee
The paper introduces a gradient-enhanced spline dimensional decomposition (GE‑SDD) surrogate that incorporates both function values and partial derivatives to improve uncertainty q…
A plug-and-play approach with fast uncertainty quantification for weak lensing mass mapping
Hubert Leterme, Andreas Tersenov, Jalal Fadili +1
The paper presents PnPMass, a plug‑and‑play algorithm that reconstructs dark‑matter maps from weak‑lensing shear data using a single deep‑learning denoiser combined with gradient d…
Maximally Robust Satisficing Bayesian Optimization
Samuli Kinnunen, Petrus Mikkola, Antti Niskanen +1
The paper proposes a Bayesian optimization method that seeks sufficiently good (satisficing) solutions which remain robust to large input perturbations after deployment, rather tha…
Predicting BESS Degradation with Uncertainty Quantification: A Probabilistic Framework for Battery Energy Storage Systems
Melina Graner, Holger Hesse, Andreas Jossen
The paper presents a deep‑learning based probabilistic framework that predicts battery state‑of‑health and quantifies uncertainty, scaling from cell‑level data to whole‑system degr…
Bayesian Inference for Extracting Barrier Distributions from Fusion Excitation Functions
Aaron Philip, Pablo Giuliani, Kyle Godbey
The paper presents a Bayesian machine‑learning approach (AutoBNN) for extracting nuclear barrier distributions from sparse fusion excitation function data, providing calibrated unc…
Neural Architectures for Amortized Bayesian Inference: Statistical Foundations and Empirical Assessments
Roy Shivam Ram Shreshtth, Arnab Hazra, Gourab Mukherjee
The paper examines how neural network architectures such as feedforward nets, Deep Sets, and Transformers can be used to amortize Bayesian inference, providing fast approximate pos…
Unified Uncertainty Quantification Framework Bridging Noisy Quantum Backends Across Variational Quantum Algorithms and Quantum Signal Processing
Priyabrata Senapati, Vibin Abraham, Qiang Guan +1
The paper introduces a unified uncertainty quantification framework that benchmarks noisy quantum processors using both variational quantum algorithms and quantum signal processing…
Variational Inference for Evidential Deep Learning
Jiawei Tang, Xinyan Du, Hui Liu +2
The paper introduces VI-EDL, a variational inference framework for evidential deep learning that controls evidence growth and provides theoretical guarantees for uncertainty estima…
Inherent interpretability provides inherent value in quantum machine learning
Kaitlin Gili, Zachary P. Bradshaw
The paper argues that the intrinsic mathematical structure of quantum machine learning models can provide inherent interpretability, offering value beyond raw performance, and illu…