paper-with-me

홈 › Papers

Training Two-Layer ReLU Networks with Gradient Descent is Inconsistent

2020-02-12 · David Holzmüller, Ingo Steinwart

We prove that two-layer (Leaky)ReLU networks initialized by e.g. the widely used method proposed by He et al. (2015) and trained using gradient descent on a least-squares loss are not universally consistent. Specifically, we describe a large class of one-dimensional data-generating distributions for which, with high probability, gradient descent only finds a bad local minimum of the optimization landscape, since it is unable to move the biases far away from their initialization at zero. It turns out that in these cases, the found network essentially performs linear regression even if the target function is non-linear. We further provide numerical evidence that this happens in practical situations, for some multi-dimensional distributions and that stochastic gradient descent exhibits similar behavior. We also provide empirical results on how the choice of initialization and optimizer can influence this behavior.

📄 PDF Abstract BibTeX arXiv:2002.04861

Code (3)

dholzmueller/nn_inconsistency 공식 구현 pytorch
MindSpore-scientific/code-2/tree/main/Two-Layer-ReLU-Network-Analytically mindspore
pwc-1/Paper-9/tree/main/4/Two-Layer-ReLU-Network-Analytically mindspore

Tasks

Vocal Bursts Valence Prediction

Methods 이 논문이 사용한 방법론

Linear Regression Linear Regression is a method for modelling a relationship between a dependent variable and independent variables. These models can be fit with numerous approaches. The most…

Similar Papers 제목 키워드 기반

Implicit Bias of Gradient Descent for Two-layer ReLU and Leaky ReLU Networks on Nearly-orthogonal Data

2023-10-29 · NeurIPS 2023 11

The implicit bias towards solutions with favorable properties is believed to be a key reason why neural networks trained by gradient-based optimization can generalize well. While the implicit bias of gradient flow has be…

Training a Two Layer ReLU Network Analytically

2023-04-06 · Adrian Barbu

Neural networks are usually trained with different variants of gradient descent based optimization algorithms such as stochastic gradient descent or the Adam optimizer. Recent theoretical work states that the critical po…

Vocal Bursts Valence Prediction

Learning One-hidden-layer ReLU Networks via Gradient Descent

2018-06-20 · Xiao Zhang, Yaodong Yu, Lingxiao Wang, Quanquan Gu

We study the problem of learning one-hidden-layer neural networks with Rectified Linear Unit (ReLU) activation function, where the inputs are sampled from standard Gaussian distribution and the outputs are generated from…

On the Convergence of Gradient Descent Training for Two-layer ReLU-networks in the Mean Field Regime

2020-05-27 · Stephan Wojtowytsch

We describe a necessary and sufficient condition for the convergence to minimum Bayes risk when training two-layer ReLU-networks by gradient descent in the mean field regime with omni-directional initial parameter distri…

Limitations of neural network training due to numerical instability of backpropagation

2022-10-03 · Clemens Karner, Vladimir Kazeev, Philipp Christian Petersen

We study the training of deep neural networks by gradient descent where floating-point arithmetic is used to compute the gradients. In this framework and under realistic assumptions, we demonstrate that it is highly unli…