中文
相关论文

相关论文: Understanding Global Loss Landscape of One-hidden-…

200 篇论文

Modern neural networks are often quite wide, causing large memory and computation costs. It is thus of great interest to train a narrower network. However, training narrow neural nets remains a challenging task. We ask two theoretical…

机器学习 · 计算机科学 2022-10-24 Jiawei Zhang , Yushun Zhang , Mingyi Hong , Ruoyu Sun , Zhi-Quan Luo

We propose ReDense as a simple and low complexity way to improve the performance of trained neural networks. We use a combination of random weights and rectified linear unit (ReLU) activation function to add a ReLU dense (ReDense) layer to…

机器学习 · 计算机科学 2020-10-27 Alireza M. Javid , Sandipan Das , Mikael Skoglund , Saikat Chatterjee

In this paper, we consider regression problems with one-hidden-layer neural networks (1NNs). We distill some properties of activation functions that lead to $\mathit{local~strong~convexity}$ in the neighborhood of the ground-truth…

机器学习 · 计算机科学 2017-06-13 Kai Zhong , Zhao Song , Prateek Jain , Peter L. Bartlett , Inderjit S. Dhillon

Recurrent Neural Networks (RNNs) are very successful at solving challenging problems with sequential data. However, this observed efficiency is not yet entirely explained by theory. It is known that a certain class of multiplicative RNNs…

机器学习 · 计算机科学 2019-01-31 Valentin Khrulkov , Oleksii Hrinchuk , Ivan Oseledets

Neural networks are playing a crucial role in everyday life, with the most modern generative models able to achieve impressive results. Nonetheless, their functioning is still not very clear, and several strategies have been adopted to…

微分几何 · 数学 2024-04-10 Alessandro Benfenati , Alessio Marta

Studying the interplay between the geometry of the loss landscape and the optimization trajectories of simple neural networks is a fundamental step for understanding their behavior in more complex settings. This paper reveals the presence…

机器学习 · 计算机科学 2024-11-08 Marco Nurisso , Pierrick Leroy , Francesco Vaccarino

Deep artificial neural networks achieve surprising generalization abilities that remain poorly understood. In this paper, we present a new approach to analyzing generalization for deep feed-forward ReLU networks that takes advantage of the…

机器学习 · 计算机科学 2023-07-06 Ramchandran Muthukumar , Jeremias Sulam

Large number of ReLU and MAC operations of Deep neural networks make them ill-suited for latency and compute-efficient private inference. In this paper, we present a model optimization method that allows a model to learn to be shallow. In…

机器学习 · 计算机科学 2023-04-27 Souvik Kundu , Yuke Zhang , Dake Chen , Peter A. Beerel

Deep neural networks' remarkable ability to correctly fit training data when optimized by gradient-based algorithms is yet to be fully understood. Recent theoretical results explain the convergence for ReLU networks that are wider than…

机器学习 · 计算机科学 2021-02-09 Asaf Noy , Yi Xu , Yonathan Aflalo , Lihi Zelnik-Manor , Rong Jin

A possible path to the interpretability of neural networks is to (approximately) represent them in the regional format of piecewise linear functions, where regions of inputs are associated to linear functions computing the network outputs.…

计算机科学中的逻辑 · 计算机科学 2025-06-09 Sandro Preto , Marcelo Finger

We study regularized deep neural networks (DNNs) and introduce a convex analytic framework to characterize the structure of the hidden layers. We show that a set of optimal hidden layer weights for a norm regularized DNN training problem…

机器学习 · 计算机科学 2021-06-14 Tolga Ergen , Mert Pilanci

Recent work has noted that all bad local minima can be removed from neural network loss landscapes, by adding a single unit with a particular parameterization. We show that the core technique from these papers can be used to remove all bad…

机器学习 · 统计学 2019-01-15 Jascha Sohl-Dickstein , Kenji Kawaguchi

The implicit bias of neural networks has been extensively studied in recent years. Lyu and Li [2019] showed that in homogeneous networks trained with the exponential or the logistic loss, gradient flow converges to a KKT point of the max…

机器学习 · 计算机科学 2022-10-04 Gal Vardi , Ohad Shamir , Nathan Srebro

Many supervised machine learning methods are naturally cast as optimization problems. For prediction models which are linear in their parameters, this often leads to convex problems for which many mathematical guarantees exist. Models which…

机器学习 · 计算机科学 2021-10-18 Francis Bach , Lenaïc Chizat

The loss surface of deep neural networks has recently attracted interest in the optimization and machine learning communities as a prime example of high-dimensional non-convex problem. Some insights were recently gained using spin glass…

机器学习 · 统计学 2017-06-05 C. Daniel Freeman , Joan Bruna

Single layer feedforward networks with random weights are known for their non-iterative and fast training algorithms and are successful in a variety of classification and regression problems. A major drawback of these networks is that they…

机器学习 · 计算机科学 2020-09-25 Ajay M. Patrikar

Neural networks have shown high successful performance in a wide range of tasks, but further studies are needed to improve its performance. We analyze the approximation error of the specific neural network architecture with a local…

机器学习 · 计算机科学 2020-09-04 Jae-Mo Kang , Sunghwan Moon

Beyond the traditional neural network training methods based on gradient descent and its variants, state estimation techniques have been proposed to determine a set of ideal weights from a control-theoretic perspective. Hence, the concept…

系统与控制 · 电气工程与系统科学 2025-08-29 Yi Yang , Victor G. Lopez , Matthias A. Müller

A ReLU neural network determines/is a continuous piecewise linear map from an input space to an output space. The weights in the neural network determine a decomposition of the input space into convex polytopes and on each of these…

机器学习 · 计算机科学 2020-12-04 Ben Sattelberg , Renzo Cavalieri , Michael Kirby , Chris Peterson , Ross Beveridge

Neural networks with random weights appear in a variety of machine learning applications, most prominently as the initialization of many deep learning algorithms and as a computationally cheap alternative to fully learned neural networks.…

机器学习 · 计算机科学 2022-11-29 Sjoerd Dirksen , Martin Genzel , Laurent Jacques , Alexander Stollenwerk