English
Related papers

Related papers: Neural (Tangent Kernel) Collapse

200 papers

How initialization and loss function affect the learning of a deep neural network (DNN), specifically its generalization error, is an important problem in practice. In this work, by exploiting the linearity of DNN training dynamics in the…

Machine Learning · Computer Science 2020-06-02 Yaoyu Zhang , Zhi-Qin John Xu , Tao Luo , Zheng Ma

A recent line of work has established intriguing connections between the generalization/compression properties of a deep neural network (DNN) model and the so-called layer weights' stable ranks. Intuitively, the latter are indicators of the…

Machine Learning · Computer Science 2021-10-07 Bogdan Georgiev , Lukas Franken , Mayukh Mukherjee , Georgios Arvanitidis

We analyze the generalization properties of two-layer neural networks in the neural tangent kernel (NTK) regime, trained with gradient descent (GD). For early stopped GD we derive fast rates of convergence that are known to be minimax…

Machine Learning · Statistics 2023-09-18 Mike Nguyen , Nicole Mücke

We prove the precise scaling, at finite depth and width, for the mean and variance of the neural tangent kernel (NTK) in a randomly initialized ReLU network. The standard deviation is exponential in the ratio of network depth to width.…

Machine Learning · Computer Science 2019-09-16 Boris Hanin , Mihai Nica

Deep neural networks (DNNs) are powerful tools for compressing and distilling information. Their scale and complexity, often involving billions of inter-dependent parameters, render direct microscopic analysis difficult. Under such…

Machine Learning · Statistics 2022-09-26 Inbar Seroussi , Gadi Naveh , Zohar Ringel

Advanced deep neural networks (DNNs), designed by either human or AutoML algorithms, are growing increasingly complex. Diverse operations are connected by complicated connectivity patterns, e.g., various types of skip connections. Those…

Machine Learning · Computer Science 2022-10-13 Wuyang Chen , Wei Huang , Xinyu Gong , Boris Hanin , Zhangyang Wang

Neural collapse is an emergent phenomenon in deep learning that was recently discovered by Papyan, Han and Donoho. We propose a simple "unconstrained features model" in which neural collapse also emerges empirically. By studying this model,…

Machine Learning · Computer Science 2020-11-24 Dustin G. Mixon , Hans Parshall , Jianzong Pi

The Neural Tangent Kernel (NTK) has emerged as a fundamental concept in the study of wide Neural Networks. In particular, it is known that the positivity of the NTK is directly related to the memorization capacity of sufficiently wide…

Machine Learning · Computer Science 2024-04-22 Luís Carvalho , João L. Costa , José Mourão , Gonçalo Oliveira

Neural networks trained with gradient descent can undergo non-trivial phase transitions as a function of the learning rate. In \cite{lewkowycz2020large} it was discovered that wide neural nets can exhibit a catapult phase for super-critical…

Machine Learning · Computer Science 2025-09-26 David Meltzer , Min Chen , Junyu Liu

Neural collapse is a highly symmetric geometric pattern of neural networks that emerges during the terminal phase of training, with profound implications on the generalization performance and robustness of the trained networks. To…

Machine Learning · Computer Science 2022-04-26 Wenlong Ji , Yiping Lu , Yiliang Zhang , Zhun Deng , Weijie J. Su

Although deep neural networks achieve tremendous success on various classification tasks, the generalization ability drops sheer when training datasets exhibit long-tailed distributions. One of the reasons is that the learned…

Machine Learning · Computer Science 2023-02-27 Xuantong Liu , Jianfeng Zhang , Tianyang Hu , He Cao , Lujia Pan , Yuan Yao

Recent work has uncovered a striking phenomenon in large-capacity neural networks: they contain blocks of contiguous hidden layers with highly similar representations. This block structure has two seemingly contradictory properties: on the…

Machine Learning · Computer Science 2022-02-16 Thao Nguyen , Maithra Raghu , Simon Kornblith

Deep learning has achieved many breakthroughs in modern classification tasks. Numerous architectures have been proposed for different data structures but when it comes to the loss function, the cross-entropy loss is the predominant choice.…

Machine Learning · Statistics 2021-12-08 Tianyang Hu , Jun Wang , Wenjia Wang , Zhenguo Li

We show that deep ensembles become equivariant for all inputs and at all training times by simply using data augmentation. Crucially, equivariance holds off-manifold and for any architecture in the infinite width limit. The equivariance is…

Machine Learning · Computer Science 2024-06-18 Jan E. Gerken , Pan Kessel

Recently, neural networks utilizing periodic activation functions have been proven to demonstrate superior performance in vision tasks compared to traditional ReLU-activated networks. However, there is still a limited understanding of the…

Machine Learning · Computer Science 2024-02-08 Hemanth Saratchandran , Shin-Fang Chng , Simon Lucey

The NTK is a widely used tool in the theoretical analysis of deep learning, allowing us to look at supervised deep neural networks through the lenses of kernel regression. Recently, several works have investigated kernel models for…

Machine Learning · Computer Science 2025-05-06 Maximilian Fleissner , Gautham Govind Anil , Debarghya Ghoshdastidar

A key property of deep neural networks (DNNs) is their ability to learn new features during training. This intriguing aspect of deep learning stands out most clearly in recently reported Grokking phenomena. While mainly reflected as a…

Machine Learning · Statistics 2024-05-07 Noa Rubin , Inbar Seroussi , Zohar Ringel

Small generalization errors of over-parameterized neural networks (NNs) can be partially explained by the frequency biasing phenomenon, where gradient-based algorithms minimize the low-frequency misfit before reducing the high-frequency…

Machine Learning · Computer Science 2022-09-27 Annan Yu , Yunan Yang , Alex Townsend

Recent research shows that for training with $\ell_2$ loss, convolutional neural networks (CNNs) whose width (number of channels in convolutional layers) goes to infinity correspond to regression with respect to the CNN Gaussian Process…

Machine Learning · Computer Science 2019-11-05 Zhiyuan Li , Ruosong Wang , Dingli Yu , Simon S. Du , Wei Hu , Ruslan Salakhutdinov , Sanjeev Arora

Deep residual network architectures have been shown to achieve superior accuracy over classical feed-forward networks, yet their success is still not fully understood. Focusing on massively over-parameterized, fully connected residual…

Machine Learning · Computer Science 2021-04-08 Yuval Belfer , Amnon Geifman , Meirav Galun , Ronen Basri