paper-with-me

Papers

Uncertainty Quantification with Bayesian Higher Order ReLU KANs

2024-10-02 · James Giroux, Cristiano Fanelli

We introduce the first method of uncertainty quantification in the domain of Kolmogorov-Arnold Networks, specifically focusing on (Higher Order) ReLUKANs to enhance computational efficiency given the computational demands of Bayesian methods. The method we propose is general in nature, providing access to both epistemic and aleatoric uncertainties. It is also capable of generalization to other various basis functions. We validate our method through a series of closure tests, including simple one-dimensional functions and application to the domain of (Stochastic) Partial Differential Equations. Referring to the latter, we demonstrate the method's ability to correctly identify functional dependencies introduced through the inclusion of a stochastic term. The code supporting this work can be found at https://github.com/wmdataphys/Bayesian-HR-KAN

📄 PDF Abstract BibTeX arXiv:2410.01687

Code (1)

wmdataphys/bayesian-hr-kan 공식 구현 pytorch

Tasks

Computational EfficiencyKolmogorov-Arnold NetworksUncertainty Quantification

Similar Papers 제목 키워드 기반

Uncertainty Quantification for Sparse Deep Learning

2020-02-26 · Yuexi Wang, Veronika Ročková

Deep learning methods continue to have a decided impact on machine learning, both in theory and in practice. Statistical theoretical developments have been mostly concerned with approximability or rates of estimation whe…

Deep LearningUncertainty Quantificationvalid

MP-GELU Bayesian Neural Networks: Moment Propagation by GELU Nonlinearity

2022-11-24 · Yuki Hirayama, Sinya Takamaeda-Yamazaki

Bayesian neural networks (BNNs) have been an important framework in the study of uncertainty quantification. Deterministic variational inference, one of the inference methods, utilizes moment propagation to compute the p…

Uncertainty QuantificationVariational Inference

The Discriminative Jackknife: Quantifying Uncertainty in Deep Learning via Higher-Order Influence Functions

2019-09-25 · Ahmed M. Alaa, Mihaela van der Schaar

Deep learning models achieve high predictive accuracy in a broad spectrum of tasks, but rigorously quantifying their predictive uncertainty remains challenging. Usable estimates of predictive uncertainty should (1) cover…

Uncertainty Quantification

Partial Trace-Class Bayesian Neural Networks

2025-11-03 · Arran Carter, Torben Sell arxiv

Bayesian neural networks (BNNs) allow rigorous uncertainty quantification in deep learning, but often come at a prohibitive computational cost. We propose three different innovative architectures of partial trace-class B…

Uncertainty-Aware Sparse Identification of Dynamical Systems via Bayesian Model Averaging

2026-04-12 · Shuhei Kashiwamura, Yusuke Kato, Hiroshi Kori, Masato Okada arxiv

In many problems of data-driven modeling for dynamical systems, the governing equations are not known a priori and must be selected phenomenologically from a large set of candidate interactions and basis functions. In su…