Newest papers

Tractable Inference with Kolmogorov-Arnold Networks

Juan Diego Toscano, Sascha Ranftl, Zhiwei Gao, Li-Lian Wang, George Em Karniadakis · 2026 · arXiv

DOI: 10.48550/arXiv.2610.29909 · open-access PDF

Korean

컴퓨터 과학의 기계 학습 대리인은 종종 정확한 예측뿐만 아니라 예상, 변수, 보존된 통합자 및 표준화 일정뿐만 아니라 필요합니다.Kolmogorov-Arnold 네트워크 (KANs)를 포함한 전통적인 신경 네트워크는 일반적으로 Monte Carlo 샘플링 또는 다차원 사각형을 통해 이러한 양을 평가하여 반복적인 추론을 비용이 많이 드는, 특히 높은 차원에서.

Machine translation (Meta M2M100); the English original follows.

Abstract

Machine-learning surrogates in computational science often require not only accurate predictions but also expectations, variances, conserved integrals, and normalization constants. Conventional neural networks, including Kolmogorov--Arnold networks (KANs), generally evaluate these quantities through Monte Carlo sampling or multidimensional quadrature, making repeated inference costly, particularly in high dimensions. We introduce KANs with Tractable Operators for Uncertainty and Constraints (TOUCKANs), whose finite-degree polynomial outer functions yield a decomposable sum--product representat

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange