Order selection with confidence for finite mixture models

Autor: Hien D. Nguyen, Daniel Fryer, Geoffrey J. McLachlan

Publikováno v: Journal of the Korean Statistical Society. 52:154-184

The determination of the number of mixture components (the order) of a finite mixture model has been an enduring problem in statistical inference. We prove that the closed testing principle leads to a sequential testing procedure (STP) that allows fo

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3d7396cb4d1f4fc74ffb1da989ce7c0b
https://doi.org/10.1007/s42952-022-00195-z

Trading performance for memory in sparse direct solvers using low-rank compression

Autor: Grégoire PICHON, Thibault Marette, Loris Marchal, Frédéric Vivien

Publikováno v: Future Generation Computer Systems
Future Generation Computer Systems, 2022, 130, pp.307-320. ⟨10.1016/j.future.2021.12.018⟩
[Research Report] RR-9368, INRIA. 2020
Future Generation Computer Systems, Elsevier, In press

Sparse direct solvers using Block Low-Rank compression have been proven efficient to solve problems arising in many real-life applications. Improving those solvers is crucial for being able to 1) solve larger problems and 2) speed up computations. A

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f63db4c7540f15de729501627227b184
https://doi.org/10.1016/j.future.2021.12.018

Two-Layer Neural Networks with Values in a Banach Space

Autor: Yury Korolev

Publikováno v: Korolev, Y 2022, ' Two-layer neural networks with values in a Banach space ', SIAM Journal on Mathematical Analysis, vol. 54, no. 6, pp. 6358-6389 . https://doi.org/10.1137/21M1458144

We study two-layer neural networks whose domain and range are Banach spaces with separable preduals. In addition, we assume that the image space is equipped with a partial order, i.e. it is a Riesz space. As the nonlinearity we choose the lattice ope

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::29a1a87b3dea86eeb4339347f289aa4d
https://doi.org/10.1137/21m1458144

A deterministic gradient-based approach to avoid saddle points

Autor: L. M. Kreusser, S. J. Osher, B. Wang

Publikováno v: Kreusser, L M, Osher, S J & Wang, B 2023, ' A Deterministic Gradient-Based Approach to Avoid Saddle Points ', European Journal of Applied Mathematics, vol. 34, no. 4, pp. 738-757 . https://doi.org/10.1017/S0956792522000316

Loss functions with a large number of saddle points are one of the major obstacles for training modern machine learning (ML) models efficiently. First-order methods such as gradient descent (GD) are usually the methods of choice for training ML model

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::138ae8b4831bd0df4615e38acd89b9c9
https://doi.org/10.1017/s0956792522000316

Time-explicit Hybrid High-Order method for the nonlinear acoustic wave equation

Autor: Steins, Morgane, Ern, Alexandre, Jamond, Olivier, Drui, Florence

We devise a fully explicit scheme for the nonlinear acoustic wave equation in its second-order formulation in time, using the HHO method for space discretization and the leapfrog scheme for time integration. The key idea for the explicitation is an i

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::d4a03cdc590e584c39ed5940b667ceac
https://hal.science/hal-03960273