paper-with-me

홈 › Papers

Towards Quantifying Intrinsic Generalization of Deep ReLU Networks

2019-10-18 · Shaeke Salman, Canlin Zhang, Xiuwen Liu, Washington Mio

Understanding the underlying mechanisms that enable the empirical successes of deep neural networks is essential for further improving their performance and explaining such networks. Towards this goal, a specific question is how to explain the "surprising" behavior of the same over-parametrized deep neural networks that can generalize well on real datasets and at the same time "memorize" training samples when the labels are randomized. In this paper, we demonstrate that deep ReLU networks generalize from training samples to new points via piece-wise linear interpolation. We provide a quantified analysis on the generalization ability of a deep ReLU network: Given a fixed point $\mathbf{x}$ and a fixed direction in the input space $\mathcal{S}$, there is always a segment such that any point on the segment will be classified the same as the fixed point $\mathbf{x}$. We call this segment the $generalization \ interval$. We show that the generalization intervals of a ReLU network behave similarly along pairwise directions between samples of the same label in both real and random cases on the MNIST and CIFAR-10 datasets. This result suggests that the same interpolation mechanism is used in both cases. Additionally, for datasets using real labels, such networks provide a good approximation of the underlying manifold in the data, where the changes are much smaller along tangent directions than along normal directions. On the other hand, however, for datasets with random labels, generalization intervals along mid-lines of triangles with the same label are much smaller than those on the datasets with real labels, suggesting different behaviors along other directions. Our systematic experiments demonstrate for the first time that such deep neural networks generalize through the same interpolation and explain the differences between their performance on datasets with real and random labels.

📄 PDF Abstract BibTeX arXiv:1910.08581

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

ReLU How Do I Communicate to Expedia? How Do I Communicate to Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Live Support & Special Travel…

Similar Papers 제목 키워드 기반

A priori generalization error for two-layer ReLU neural network through minimum norm solution

2019-12-06 · Zhi-Qin John Xu, Jiwei Zhang, Yaoyu Zhang, Chengchao Zhao

We focus on estimating \emph{a priori} generalization error of two-layer ReLU neural networks (NNs) trained by mean squared error, which only depends on initial parameters and the target function, through the following r…

A rescaling-invariant Lipschitz bound based on path-metrics for modern ReLU network parameterizations

2024-05-23 · Antoine Gonon, Nicolas Brisebarre, Elisa Riccietti, Rémi Gribonval

Lipschitz bounds on neural network parameterizations are important to establish generalization, quantization or pruning guarantees, as they control the robustness of the network with respect to parameter changes. Yet, th…

Generalization BoundsNetwork PruningQuantization

EraseReLU: A Simple Way to Ease the Training of Deep Convolution Neural Networks

2017-09-22 · Xuanyi Dong, Guoliang Kang, Kun Zhan, Yi Yang

For most state-of-the-art architectures, Rectified Linear Unit (ReLU) becomes a standard component accompanied with each layer. Although ReLU can ease the network training to an extent, the character of blocking negative…

BlockingImage Classification

Convex Geometry of ReLU-layers, Injectivity on the Ball and Local Reconstruction

2023-07-18 · Daniel Haider, Martin Ehler, Peter Balazs

The paper uses a frame-theoretic setting to study the injectivity of a ReLU-layer on the closed ball of $\mathbb{R}^n$ and its non-negative part. In particular, the interplay between the radius of the ball and the bias v…

Generalization Below the Edge of Stability: The Role of Data Geometry

2025-10-20 · Tongtong Liang, Alexander Cloninger, Rahul Parhi, Yu-Xiang Wang arxiv

Understanding generalization in overparameterized neural networks hinges on the interplay between the data geometry, neural architecture, and training dynamics. In this paper, we theoretically explore how data geometry c…