paper-with-me

Papers

Information-Theoretic Generalization Bounds for Deep Neural Networks

2024-04-04 · Haiyun He, Ziv Goldfeld

Deep neural networks (DNNs) exhibit an exceptional capacity for generalization in practical applications. This work aims to capture the effect and benefits of depth for supervised learning via information-theoretic generalization bounds. We first derive two hierarchical bounds on the generalization error in terms of the Kullback-Leibler (KL) divergence or the 1-Wasserstein distance between the train and test distributions of the network internal representations. The KL divergence bound shrinks as the layer index increases, while the Wasserstein bound implies the existence of a layer that serves as a generalization funnel, which attains a minimal 1-Wasserstein distance. Analytic expressions for both bounds are derived under the setting of binary Gaussian classification with linear DNNs. To quantify the contraction of the relevant information measures when moving deeper into the network, we analyze the strong data processing inequality (SDPI) coefficient between consecutive layers of three regularized DNN models: $\mathsf{Dropout}$, $\mathsf{DropConnect}$, and Gaussian noise injection. This enables refining our generalization bounds to capture the contraction as a function of the network architecture parameters. Specializing our results to DNNs with a finite parameter space and the Gibbs algorithm reveals that deeper yet narrower network architectures generalize better in those examples, although how broadly this statement applies remains a question.

📄 PDF Abstract BibTeX arXiv:2404.03176

Code (0)

등록된 구현이 없습니다.

Tasks

Generalization Bounds

Methods 이 논문이 사용한 방법론

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
DropConnect DropConnect generalizes Dropout by randomly dropping the weights rather than the activations with probability $1-p$. DropConnect…

Similar Papers 제목 키워드 기반

Formal limitations of sample-wise information-theoretic generalization bounds

2022-05-13 · Hrayr Harutyunyan, Greg Ver Steeg, Aram Galstyan

Some of the tightest information-theoretic generalization bounds depend on the average information between the learned hypothesis and a single training example. However, these sample-wise bounds were derived only for exp…

Generalization Bounds

Information Theoretic Lower Bounds for Information Theoretic Upper Bounds

2023-02-09 · NeurIPS 2023 11 · Roi Livni

We examine the relationship between the mutual information between the output model and the empirical sample and the generalization of the algorithm in the context of stochastic convex optimization. Despite increasing in…

Generalization Bounds

Information-theoretic generalization bounds for black-box learning algorithms

2021-10-04 · NeurIPS 2021 12 · Hrayr Harutyunyan, Maxim Raginsky, Greg Ver Steeg, Aram Galstyan

We derive information-theoretic generalization bounds for supervised learning algorithms based on the information contained in predictions rather than in the output of the training algorithm. These bounds improve over th…

Generalization Bounds

An Information-Theoretic Framework for Out-of-Distribution Generalization with Applications to Stochastic Gradient Langevin Dynamics

2024-03-29 · Wenliang Liu, Guanding Yu, Lele Wang, Renjie Liao

We study the Out-of-Distribution (OOD) generalization in machine learning and propose a general framework that establishes information-theoretic generalization bounds. Our framework interpolates freely between Integral P…

Generalization BoundsOut-of-Distribution Generalization

Sample-Conditioned Hypothesis Stability Sharpens Information-Theoretic Generalization Bounds

2023-10-31 · NeurIPS 2023 11

We present new information-theoretic generalization guarantees through the a novel construction of the "neighboring-hypothesis" matrix and a new family of stability notions termed sample-conditioned hypothesis (SCH) stab…

Generalization Bounds