中文
相关论文

相关论文: How Uniform Random Weights Induce Non-uniform Bias…

200 篇论文

Systematic generalization aims to evaluate reasoning about novel combinations from known components, an intrinsic property of human cognition. In this work, we study systematic generalization of NNs in forecasting future time series of…

机器学习 · 计算机科学 2021-03-09 Hritik Bansal , Gantavya Bhatt , Pankaj Malhotra , Prathosh A. P

Within the machine learning community, the widely-used uniform convergence framework has been used to answer the question of how complex, over-parameterized models can generalize well to new data. This approach bounds the test error of the…

机器学习 · 统计学 2021-03-05 Ryan Theisen , Jason M. Klusowski , Michael W. Mahoney

Existing generalization measures that aim to capture a model's simplicity based on parameter counts or norms fail to explain generalization in overparameterized deep neural networks. In this paper, we introduce a new, theoretically…

机器学习 · 计算机科学 2021-03-11 Lorenz Kuhn , Clare Lyle , Aidan N. Gomez , Jonas Rothfuss , Yarin Gal

Deep learning relies on good initialization schemes and hyperparameter choices prior to training a neural network. Random weight initializations induce random network ensembles, which give rise to the trainability, training speed, and…

机器学习 · 统计学 2019-10-25 Rebekka Burkholz , Alina Dubatovka

Mini-batch training of Graph Neural Networks (GNNs) is fundamentally different from training on i.i.d. data: sampling a subgraph alters the topology and introduces boundary effects, leading prior work to develop structure-aware samplers…

机器学习 · 计算机科学 2026-05-22 Clement Wang , Antoine Vialle , Robin Vaysse , Thomas Bonald

Overparameterized neural networks (NNs) are observed to generalize well even when trained to perfectly fit noisy data. This phenomenon motivated a large body of work on "benign overfitting", where interpolating predictors achieve…

机器学习 · 计算机科学 2024-03-22 Guy Kornowski , Gilad Yehudai , Ohad Shamir

We propose tensorial neural networks (TNNs), a generalization of existing neural networks by extending tensor operations on low order operands to those on high order ones. The problem of parameter learning is challenging, as it corresponds…

机器学习 · 统计学 2018-12-11 Jiahao Su , Jingling Li , Bobby Bhattacharjee , Furong Huang

It is well known that direct training of deep neural networks will generally lead to poor results. A major progress in recent years is the invention of various pretraining methods to initialize network parameters and it was shown that such…

机器学习 · 计算机科学 2014-06-10 Jun Li , Wei Luo , Jian Yang , Xiaotong Yuan

Due to the over-parameterization nature, neural networks are a powerful tool for nonlinear function approximation. In order to achieve good generalization on unseen data, a suitable inductive bias is of great importance for neural networks.…

机器学习 · 计算机科学 2021-11-17 Weiyang Liu , Rongmei Lin , Zhen Liu , Li Xiong , Bernhard Schölkopf , Adrian Weller

We examine the necessity of interpolation in overparameterized models, that is, when achieving optimal predictive risk in machine learning problems requires (nearly) interpolating the training data. In particular, we consider simple…

机器学习 · 统计学 2022-06-17 Chen Cheng , John Duchi , Rohith Kuditipudi

This paper analyzes the convergence and generalization of training a one-hidden-layer neural network when the input features follow the Gaussian mixture model consisting of a finite number of Gaussian distributions. Assuming the labels are…

机器学习 · 计算机科学 2023-01-30 Hongkang Li , Shuai Zhang , Meng Wang

Despite Deep Learning's (DL) empirical success, our theoretical understanding of its efficacy remains limited. One notable paradox is that while conventional wisdom discourages perfect data fitting, deep neural networks are designed to do…

机器学习 · 计算机科学 2024-02-06 Oria Gruber , Haim Avron

Graph Neural Networks (GNNs) often perform better for high-degree nodes than low-degree nodes on node classification tasks. This degree bias can reinforce social marginalization by, e.g., privileging celebrities and other high-degree actors…

机器学习 · 计算机科学 2024-11-04 Arjun Subramonian , Jian Kang , Yizhou Sun

Deep neural networks are typically trained by optimizing a loss function with an SGD variant, in conjunction with a decaying learning rate, until convergence. We show that simple averaging of multiple points along the trajectory of SGD,…

机器学习 · 计算机科学 2019-02-26 Pavel Izmailov , Dmitrii Podoprikhin , Timur Garipov , Dmitry Vetrov , Andrew Gordon Wilson

Neural networks have attracted a lot of attention due to its success in applications such as natural language processing and computer vision. For large scale data, due to the tremendous number of parameters in neural networks, overfitting…

机器学习 · 统计学 2022-07-05 Xiaoxi Shen , Jinghang Lin

Recurrent and deep neural networks (RNNs/DNNs) are cornerstone architectures in machine learning. Remarkably, RNNs differ from DNNs only by weight sharing, as can be shown through unrolling in time. How does this structural similarity fit…

机器学习 · 计算机科学 2026-02-18 Jan P. Bauer , Kirsten Fischer , Moritz Helias , Agostina Palmigiano

Standard deep neural networks (DNNs) are commonly trained in an end-to-end fashion for specific tasks such as object recognition, face identification, or character recognition, among many examples. This specificity often leads to…

计算机视觉与模式识别 · 计算机科学 2020-07-14 Raphaël Achddou , J. Matias di Martino , Guillermo Sapiro

There is a significant performance gap between Binary Neural Networks (BNNs) and floating point Deep Neural Networks (DNNs). We propose to improve the binary training method, by introducing a new regularization function that encourages…

机器学习 · 计算机科学 2020-04-22 Sajad Darabi , Mouloud Belbahri , Matthieu Courbariaux , Vahid Partovi Nia

We carry out an information-theoretical analysis of a two-layer neural network trained from input-output pairs generated by a teacher network with matching architecture, in overparametrized regimes. Our results come in the form of bounds…

机器学习 · 计算机科学 2023-07-13 Francesco Camilli , Daria Tieplova , Jean Barbier

Neural networks are known to develop latent representations that are $aligned$, namely structurally similar across networks trained with different architectures, training protocols, or training datasets. We study this phenomenon in a…

机器学习 · 统计学 2026-05-27 Ali Hussaini Umar , Alessandro Laio
‹ 上一页 1 8 9 10 下一页 ›