paper-with-me

Papers

How Does the ReLU Activation Affect the Implicit Bias of Gradient Descent on High-dimensional Neural Network Regression?

2026-03-05 · Kuo-Wei Lai, Guanghui Wang, Molei Tao, Vidya Muthukumar arxiv

Overparameterized ML models, including neural networks, typically induce underdetermined training objectives with multiple global minima. The implicit bias refers to the limiting global minimum that is attained by a common optimization algorithm, such as gradient descent (GD). In this paper, we characterize the implicit bias of GD for training a shallow ReLU model with the squared loss on high-dimensional random features. Prior work (Vardi and Shamir, 2021) showed that the implicit bias does not exist in the worst-case, or corresponds exactly to the minimum-$\ell_2$-norm interpolating solution under exactly orthogonal data (Boursier et al., 2022). Our work interpolates between these two extremes and shows that, for sufficiently high-dimensional random data, the implicit bias approximates the minimum-$\ell_2$-norm solution with high probability with a gap on the order $Θ(\sqrt{n/||λ||_1})$, where $n$ is the number of training examples and $λ$ denotes the spectrum of the data covariance matrix. Our results are obtained through a novel primal-dual analysis that carefully tracks the evolution of predictions, data-span coefficients, as well as their interactions, and show that the ReLU activation pattern quickly stabilizes with high probability over random data.

📄 PDF Abstract BibTeX arXiv:2603.04895

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Do We Always Need the Simplicity Bias? Looking for Optimal Inductive Biases in the Wild

2025-03-13 · CVPR 2025 1 · Damien Teney, Liangze Jiang, Florin Gogianu, Ehsan Abbasnejad

Neural architectures tend to fit their data with relatively simple functions. This "simplicity bias" is widely regarded as key to their success. This paper explores the limits of this principle. Building on recent findin…

image-classificationImage Classification

Implicit Bias of Gradient Descent for Two-layer ReLU and Leaky ReLU Networks on Nearly-orthogonal Data

2023-10-29 · NeurIPS 2023 11

The implicit bias towards solutions with favorable properties is believed to be a key reason why neural networks trained by gradient-based optimization can generalize well. While the implicit bias of gradient flow has be…

Can Implicit Bias Imply Adversarial Robustness?

2024-05-24 · Hancheng Min, René Vidal

The implicit bias of gradient-based training algorithms has been considered mostly beneficial as it leads to trained networks that often generalize well. However, Frei et al. (2023) show that such implicit bias can harm …

Adversarial Robustness

ReLUs Are Sufficient for Learning Implicit Neural Representations

2024-06-04 · Joseph Shenouda, Yamin Zhou, Robert D. Nowak

Motivated by the growing theoretical understanding of neural networks that employ the Rectified Linear Unit (ReLU) as their activation function, we revisit the use of ReLU activation functions for learning implicit neura…

Super-Resolution

Implicit Bias of Gradient Descent for Wide Two-layer Neural Networks Trained with the Logistic Loss

2020-02-11 · Lenaic Chizat, Francis Bach

Neural networks trained to minimize the logistic (a.k.a. cross-entropy) loss with gradient-based methods are observed to perform well in many supervised classification tasks. Towards understanding this phenomenon, we ana…

Generalization Bounds