Related papers: Uncertainty-guided active learning for surrogate prediction of stream-finishing wear fields

Uncertainty-guided active learning for surrogate prediction of stream-finishing wear fields

URL: http://arxiv.org/abs/2608.00593v1
Date: Sat, 01 Aug 2026 11:12:39 GMT
Title: Uncertainty-guided active learning for surrogate prediction of stream-finishing wear fields
Abstract summary: We present an uncertainty-guided surrogate framework that predicts, directly from geometry, the three fields governing erosion.<n>The surrogate employs a deep ensemble whose disagreement estimates uncertainty, enabling an active-learning strategy that performs DEM simulations for the most uncertain orientations.
Score: 37.82552353322716
License: http://creativecommons.org/licenses/by/4.0/
Abstract: In stream finishing, the wear experienced by a workpiece depends strongly on its orientation within the rotating abrasive media. Determining suitable orientations to achieve uniform wear requires evaluating the wear-rate field over all feasible orientations. Although the discrete element method (DEM) accurately resolves particle interactions, simulating hundreds of feasible orientations for a new geometry is computationally expensive. We present an uncertainty-guided surrogate framework that predicts, directly from geometry, the three fields governing erosion: per-triangle normal impact velocity, tangential impact velocity, and particle impact flux. These fields are combined through the Finnie wear model to reconstruct the wear-rate distribution. The surrogate employs a deep ensemble whose disagreement estimates epistemic uncertainty, enabling an active-learning strategy that selectively performs DEM simulations for the most uncertain orientations. Trained using only $13\%$ of the $696$ feasible orientations, the surrogate achieves Spearman rank correlations of $0.93$, $0.89$, and $0.93$ for the normal impact velocity, tangential impact velocity, and particle impact flux, respectively. Moreover, the predicted uncertainty is well calibrated, reliably anticipating prediction error and the fidelity of the reconstructed wear field, which matches DEM with a Spearman rank correlation of up to $0.97$ for low-uncertainty orientations and degrades in a controlled manner as uncertainty increases.

Related papers

Equation-Free Period-Aware Forecast-Error Contraction for Estimating Negative Largest Lyapunov Exponents from Short Trajectory Ensembles [4.689984557057394]
We introduce a period-aware forecast-error contraction procedure for estimating a dominant negative Lyapunov exponent from ensembles of short scalar trajectories.<n>A k-nearest-neighbor predictor is trained on trajectory histories, the geometric-mean absolute forecast error is evaluated at phase-consistent horizons, and the exponent is obtained from the slope of the logarithmic error profile.
arXiv Detail & Related papers (2026-08-06T01:58:00Z)
The Geometric Nature and a Free Proxy for Flow-Matching Uncertainty [40.15880250907137]
Flow matching (FM) has become a popular action head paradigm for modern embodied models.<n>We show that uncertainty manifests as deviation from an ideal affine-isotropic contraction field.<n>We introduce denoising acceleration ($mathrmaccel$), a cost-free uncertainty proxy that measures the bending of the trajectory from a single forward pass.
arXiv Detail & Related papers (2026-07-30T09:44:46Z)
Beyond Point Estimates: Benchmarking Uncertainty Quantification Methods on the AION-1 Astronomical Foundation Model [47.71489969727656]
Conformalized Quantile Regression (CQR) delivers the best coverage in the bin with the poorest model predictions.<n>Only the Locally Valid and Discriminative (LVD) framework provides finite-sample emphlocal validity.
arXiv Detail & Related papers (2026-06-05T18:29:19Z)
Autonomous Adaptive Solver Selection for Chemistry Integration via Reinforcement Learning [0.0]
We introduce a constrained reinforcement learning (RL) framework that autonomously selects between an implicit BDF integrator (CVODE) and a quasi-steady-state (QSS) solver during chemistry integration.<n>Across 0D homogeneous reactor conditions, the RL-adaptive policy achieves a mean speedup of approximately $3times$, with speedups ranging from $1.11times$ to $10.58times$.<n>Without retraining, the 0D-trained policy transfers to 1D counterflow diffusion flames over strain rates $10$--$2000mathrms-1$, delivering consistent $approx 2.2
arXiv Detail & Related papers (2026-03-31T21:44:58Z)
Scalar-Measurement Attitude Estimation on $\mathbf{SO}(3)$ with Bias Compensation [35.18016233072556]
Attitude estimation methods typically rely on full vector measurements from inertial sensors such as accelerometers and magnetometers.<n>This paper shows that reliable estimation can also be achieved using only scalar measurements.<n>We propose nonlinear deterministic observers on $mathbfSO(3)$ that incorporate gyroscope bias compensation.
arXiv Detail & Related papers (2026-03-03T00:02:23Z)
Binary Flow Matching: Prediction-Loss Space Alignment for Robust Learning [23.616336786063552]
Flow matching has emerged as a powerful framework for generative modeling.<n>We identify a latent structural mismatch that arises when it is coupled with velocity-based objectives.<n>We prove that re-aligning the objective to the signal space eliminates the singular weighting.
arXiv Detail & Related papers (2026-02-11T02:02:30Z)
Equivariant Evidential Deep Learning for Interatomic Potentials [55.6997213490859]
Uncertainty quantification is critical for assessing the reliability of machine learning interatomic potentials in molecular dynamics simulations.<n>Existing UQ approaches for MLIPs are often limited by high computational cost or suboptimal performance.<n>We propose textitEquivariant Evidential Deep Learning for Interatomic Potentials ($texte2$IP), a backbone-agnostic framework that models atomic forces and their uncertainty jointly.
arXiv Detail & Related papers (2026-02-11T02:00:25Z)
Information Fidelity in Tool-Using LLM Agents: A Martingale Analysis of the Model Context Protocol [69.11739400975445]
We introduce the first theoretical framework for analyzing error accumulation in Model Context Protocol (MCP) agents.<n>We show that cumulative distortion exhibits linear growth and high-probability deviations bounded by $O(sqrtT)$.<n>Key findings include: semantic weighting reduces distortion by 80%, and periodic re-grounding approximately every 9 steps suffices for error control.
arXiv Detail & Related papers (2026-02-10T21:08:53Z)
Revisiting Zeroth-Order Optimization: Minimum-Variance Two-Point Estimators and Directionally Aligned Perturbations [57.179679246370114]
We identify the distribution of random perturbations that minimizes the estimator's variance as the perturbation stepsize tends to zero.<n>Our findings reveal that such desired perturbations can align directionally with the true gradient, instead of maintaining a fixed length.
arXiv Detail & Related papers (2025-10-22T19:06:39Z)
Statistical Learning with Sublinear Regret of Propagator Models [3.1755820123640612]
We consider a class of learning problems in which an agent liquidates a risky asset while creating both transient impact price driven by an unknown convolution propagator and linear temporary impact price with an unknown parameter.<n>We present a trading algorithm that alternates between exploration and exploitation and sublinear sublinear regrets with high probability.
arXiv Detail & Related papers (2023-01-12T17:16:27Z)
Temporal Difference Uncertainties as a Signal for Exploration [76.6341354269013]
An effective approach to exploration in reinforcement learning is to rely on an agent's uncertainty over the optimal policy. In this paper, we highlight that value estimates are easily biased and temporally inconsistent. We propose a novel method for estimating uncertainty over the value function that relies on inducing a distribution over temporal difference errors.
arXiv Detail & Related papers (2020-10-05T18:11:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.