paper-with-me

홈 › Papers

Training of deep residual networks with stochastic MG/OPT

2021-08-09 · Cyrill von Planta, Alena Kopanicakova, Rolf Krause

We train deep residual networks with a stochastic variant of the nonlinear multigrid method MG/OPT. To build the multilevel hierarchy, we use the dynamical systems viewpoint specific to residual networks. We report significant speed-ups and additional robustness for training MNIST on deep residual networks. Our numerical experiments also indicate that multilevel training can be used as a pruning technique, as many of the auxiliary networks have accuracies comparable to the original network.

📄 PDF Abstract BibTeX arXiv:2108.04052

Code (1)

eulerinstitute/mgopt_icml21 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Pruning 설명 없음

Similar Papers 제목 키워드 기반

Stochastic Training of Residual Networks: a Differential Equation Viewpoint

2018-12-01 · Qi Sun, Yunzhe Tao, Qiang Du

During the last few years, significant attention has been paid to the stochastic training of artificial neural networks, which is known as an effective regularization approach that helps improve the generalization capabi…

image-classificationImage Classification

KromHC: Manifold-Constrained Hyper-Connections with Kronecker-Product Residual Matrices

2026-01-29 · Wuyang Zhou, Yuxuan Gu, Giorgos Iacovides, Danilo Mandic arxiv

The success of Hyper-Connections (HC) in neural networks (NN) has also highlighted issues related to training instability and restricted scalability. The Manifold-Constrained Hyper-Connections (mHC) mitigate these challe…

Exploring epoch-dependent stochastic residual networks

2017-04-20 · Fabio Carrara, Andrea Esuli, Fabrizio Falchi, Alejandro Moreo Fernández

The recently proposed stochastic residual networks selectively activate or bypass the layers during training, based on independent stochastic choices, each of which following a probability distribution that is fixed in a…

Management

mHC-lite: You Don't Need 20 Sinkhorn-Knopp Iterations

2026-01-09 · Yongyi Yang, Jianyang Gao arxiv

Hyper-Connections (HC) generalizes residual connections by introducing dynamic residual matrices that mix information across multiple residual streams, accelerating convergence in deep neural networks. However, unconstra…

Mind the Residual Gap: Probabilistic Downscaling under Real-World Bias

2026-06-29 · Yujin Kim, Nidhi Soma, Sarah Dean arxiv

Probabilistic downscaling is the task of modeling the conditional distribution of high-resolution fields given coarse inputs, and is a central challenge to atmospheric science, climate modeling, and other multiscale phys…