paper-with-me

Papers

Sharpened Generalization Bounds based on Conditional Mutual Information and an Application to Noisy, Iterative Algorithms

2020-04-27 · NeurIPS 2020 12 · Mahdi Haghifam, Jeffrey Negrea, Ashish Khisti, Daniel M. Roy, Gintare Karolina Dziugaite

The information-theoretic framework of Russo and J. Zou (2016) and Xu and Raginsky (2017) provides bounds on the generalization error of a learning algorithm in terms of the mutual information between the algorithm's output and the training sample. In this work, we study the proposal, by Steinke and Zakynthinou (2020), to reason about the generalization error of a learning algorithm by introducing a super sample that contains the training sample as a random subset and computing mutual information conditional on the super sample. We first show that these new bounds based on the conditional mutual information are tighter than those based on the unconditional mutual information. We then introduce yet tighter bounds, building on the "individual sample" idea of Bu, S. Zou, and Veeravalli (2019) and the "data dependent" ideas of Negrea et al. (2019), using disintegrated mutual information. Finally, we apply these bounds to the study of Langevin dynamics algorithm, showing that conditioning on the super sample allows us to exploit information in the optimization trajectory to obtain tighter bounds based on hypothesis tests.

📄 PDF Abstract BibTeX arXiv:2004.12983

Code (0)

등록된 구현이 없습니다.

Tasks

Generalization Bounds

Similar Papers 제목 키워드 기반

Individually Conditional Individual Mutual Information Bound on Generalization Error

2020-12-17 · Ruida Zhou, Chao Tian, Tie Liu

We propose a new information-theoretic bound on generalization error based on a combination of the error decomposition technique of Bu et al. and the conditional mutual information (CMI) construction of Steinke and Zakyn…

LEMMA

PAC-Bayes, MAC-Bayes and Conditional Mutual Information: Fast rate bounds that handle general VC classes

2021-06-17 · Peter Grünwald, Thomas Steinke, Lydia Zakynthinou

We give a novel, unified derivation of conditional PAC-Bayesian and mutual information (MI) generalization bounds. We derive conditional MI bounds as an instance, with special choice of prior, of conditional MAC-Bayesian…

Generalization Bounds

On Leave-One-Out Conditional Mutual Information For Generalization

2022-07-01 · Mohamad Rida Rammal, Alessandro Achille, Aditya Golatkar, Suhas Diggavi 외

We derive information theoretic generalization bounds for supervised learning algorithms based on a new measure of leave-one-out conditional mutual information (loo-CMI). Contrary to other CMI bounds, which are black-box…

Generalization Boundsimage-classificationImage Classification

Reasoning About Generalization via Conditional Mutual Information

2020-01-24 · Thomas Steinke, Lydia Zakynthinou

We provide an information-theoretic framework for studying the generalization properties of machine learning algorithms. Our framework ties together existing approaches, including uniform convergence bounds and recent me…

BIG-bench Machine Learning

Limitations of Information-Theoretic Generalization Bounds for Gradient Descent Methods in Stochastic Convex Optimization

2022-12-27 · Mahdi Haghifam, Borja Rodríguez-Gálvez, Ragnar Thobaben, Mikael Skoglund 외

To date, no "information-theoretic" frameworks for reasoning about generalization error have been shown to establish minimax rates for gradient descent in the setting of stochastic convex optimization. In this work, we c…

Generalization Bounds