中文
相关论文

相关论文: On the Decision Boundary of Deep Neural Networks

200 篇论文

The stunning empirical successes of neural networks currently lack rigorous theoretical explanation. What form would such an explanation take, in the face of existing complexity-theoretic lower bounds? A first step might be to show that…

机器学习 · 计算机科学 2017-07-18 Le Song , Santosh Vempala , John Wilmes , Bo Xie

Efficient model selection for identifying a suitable pre-trained neural network to a downstream task is a fundamental yet challenging task in deep learning. Current practice requires expensive computational costs in model training for…

机器学习 · 计算机科学 2022-01-19 Chunheng Jiang , Tejaswini Pedapati , Pin-Yu Chen , Yizhou Sun , Jianxi Gao

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

Deep neural networks can be effective means to automatically classify aerial images but is easy to overfit to the training data. It is critical for trained neural networks to be robust to variations that exist between training and test…

计算机视觉与模式识别 · 计算机科学 2019-09-25 Jiayun Wang , Patrick Virtue , Stella X. Yu

Training deep neural networks results in strong learned representations that show good generalization capabilities. In most cases, training involves iterative modification of all weights inside the network via back-propagation. In Extreme…

机器学习 · 计算机科学 2018-02-06 Amir Rosenfeld , John K. Tsotsos

The solution to partial differential equations using deep learning approaches has shown promising results for several classes of initial and boundary-value problems. However, their ability to surpass, particularly in terms of accuracy,…

数值分析 · 数学 2023-08-23 Ziad Aldirany , Régis Cottereau , Marc Laforest , Serge Prudhomme

The largely successful method of training neural networks is to learn their weights using some variant of stochastic gradient descent (SGD). Here, we show that the solutions found by SGD can be further improved by ensembling a subset of the…

Our understanding of learning dynamics of deep neural networks (DNNs) remains incomplete. Recent research has begun to uncover the mathematical principles underlying these networks, including the phenomenon of "Neural Collapse", where…

机器学习 · 计算机科学 2024-02-13 Bradley T. Baker , Barak A. Pearlmutter , Robyn Miller , Vince D. Calhoun , Sergey M. Plis

Understanding the per-layer learning dynamics of deep neural networks is of significant interest as it may provide insights into how neural networks learn and the potential for better training regimens. We investigate learning in Deep…

机器学习 · 计算机科学 2020-12-02 Ayush Manish Agrawal , Atharva Tendle , Harshvardhan Sikka , Sahib Singh , Amr Kayid

Pre-trained Deep Neural Network (DNN) models are increasingly used in smartphones and other user devices to enable prediction services, leading to potential disclosures of (sensitive) information from training data captured inside these…

密码学与安全 · 计算机科学 2019-07-16 Fan Mo , Ali Shahin Shamsabadi , Kleomenis Katevas , Andrea Cavallaro , Hamed Haddadi

We consider the problem of selective prediction (also known as reject option) in deep neural networks, and introduce SelectiveNet, a deep neural architecture with an integrated reject option. Existing rejection mechanisms are based mostly…

机器学习 · 计算机科学 2019-06-28 Yonatan Geifman , Ran El-Yaniv

Modeling uncertainty in deep neural networks, despite recent important advances, is still an open problem. Bayesian neural networks are a powerful solution, where the prior over network weights is a design choice, often a normal…

机器学习 · 统计学 2019-10-29 Raanan Y. Rohekar , Yaniv Gurwicz , Shami Nisimov , Gal Novik

The brain processes information through many layers of neurons. This deep architecture is representationally powerful, but it complicates learning by making it hard to identify the responsible neurons when a mistake is made. In machine…

神经元与认知 · 定量生物学 2014-11-04 Timothy P. Lillicrap , Daniel Cownden , Douglas B. Tweed , Colin J. Akerman

Deep neural networks (DNN) such as Multi-Layer Perception (MLP) and Convolutional Neural Networks (CNN) represent one of the most established deep learning algorithms. Given the tremendous effects of the number of hidden layers on network…

图像与视频处理 · 电气工程与系统科学 2021-11-17 Shihao Shao , Yong Liu , Qinghua Cui

This work finds the analytical expression of the global minima of a deep linear network with weight decay and stochastic neurons, a fundamental model for understanding the landscape of neural networks. Our result implies that the origin is…

机器学习 · 统计学 2023-06-14 Liu Ziyin , Botao Li , Xiangming Meng

Among the several paradigms of artificial intelligence (AI) or machine learning (ML), a remarkably successful paradigm is deep learning. Deep learning's phenomenal success has been hoped to be interpreted via fundamental research on the…

机器学习 · 计算机科学 2021-11-29 Tilahun M. Getu

The tremendous recent progress in analyzing the training dynamics of overparameterized neural networks has primarily focused on wide networks and therefore does not sufficiently address the role of depth in deep learning. In this work, we…

机器学习 · 计算机科学 2022-06-29 Jongmin Lee , Joo Young Choi , Ernest K. Ryu , Albert No

This dissertation studies a fundamental open challenge in deep learning theory: why do deep networks generalize well even while being overparameterized, unregularized and fitting the training data to zero error? In the first part of the…

机器学习 · 计算机科学 2021-10-19 Vaishnavh Nagarajan

A longstanding goal in deep learning research has been to precisely characterize training and generalization. However, the often complex loss landscapes of neural networks have made a theory of learning dynamics elusive. In this work, we…

Predictive coding networks are neural models that perform inference through an iterative energy minimization process, whose operations are local in space and time. While effective in shallow architectures, they suffer significant…

机器学习 · 计算机科学 2025-10-13 Chang Qi , Matteo Forasassi , Thomas Lukasiewicz , Tommaso Salvatori