paper-with-me

홈 › Papers

A unified recipe for deriving (time-uniform) PAC-Bayes bounds

2023-02-07 · Ben Chugg, Hongjian Wang, Aaditya Ramdas

We present a unified framework for deriving PAC-Bayesian generalization bounds. Unlike most previous literature on this topic, our bounds are anytime-valid (i.e., time-uniform), meaning that they hold at all stopping times, not only for a fixed sample size. Our approach combines four tools in the following order: (a) nonnegative supermartingales or reverse submartingales, (b) the method of mixtures, (c) the Donsker-Varadhan formula (or other convex duality principles), and (d) Ville's inequality. Our main result is a PAC-Bayes theorem which holds for a wide class of discrete stochastic processes. We show how this result implies time-uniform versions of well-known classical PAC-Bayes bounds, such as those of Seeger, McAllester, Maurer, and Catoni, in addition to many recent bounds. We also present several novel bounds. Our framework also enables us to relax traditional assumptions; in particular, we consider nonstationary loss functions and non-i.i.d. data. In sum, we unify the derivation of past bounds and ease the search for future bounds: one may simply check if our supermartingale or submartingale conditions are met and, if so, be guaranteed a (time-uniform) PAC-Bayes bound.

📄 PDF Abstract BibTeX arXiv:2302.03421

Code (0)

등록된 구현이 없습니다.

Tasks

Generalization Boundsvalid

Similar Papers 제목 키워드 기반

A Limitation of the PAC-Bayes Framework

2020-06-24 · NeurIPS 2020 12 · Roi Livni, Shay Moran

PAC-Bayes is a useful framework for deriving generalization bounds which was introduced by McAllester ('98). This framework has the flexibility of deriving distribution- and algorithm-dependent bounds, which are often ti…

Generalization Bounds

A DPI-PAC-Bayesian Framework for Generalization Bounds

2025-07-20 · Muhan Guan, Farhad Farokhi, Jingge Zhu arxiv

We develop a unified Data Processing Inequality PAC-Bayesian framework -- abbreviated DPI-PAC-Bayesian -- for deriving generalization error bounds in the supervised learning setting. By embedding the Data Processing Ineq…

A Unified Confidence Sequence for Generalized Linear Models, with Applications to Bandits

2024-07-19 · Junghyun Lee, Se-Young Yun, Kwang-Sung Jun

We present a unified likelihood ratio-based confidence sequence (CS) for any (self-concordant) generalized linear model (GLM) that is guaranteed to be convex and numerically tight. We show that this is on par or improves…

LEMMA

Tighter Generalisation Bounds via Interpolation

2024-02-07 · Paul Viallard, Maxime Haddouche, Umut Şimşekli, Benjamin Guedj

This paper contains a recipe for deriving new PAC-Bayes generalisation bounds based on the $(f, \Gamma)$-divergence, and, in addition, presents PAC-Bayes generalisation bounds where we interpolate between a series of pro…

Rethinking Shrinkage Bias in LLM FP4 Pretraining: Geometric Origin, Systemic Impact, and UFP4 Recipe

2026-06-18 · Qian Zhao, Kunlong Chen, Changxin Tian, Zhonghui Jiang 외 arxiv

FP4 training promises substantial reductions in memory and computation cost for LLM pretraining, yet current FP4 hardware paths and recipes, including NVIDIA Blackwell/Rubin-class systems and AMD MI350-series GPUs, remai…