arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.06589cs.LGcs.AIstat.ML
收藏

基于PAC-Bayes和一致稳定性的元学习推广界

Generalization Bounds for Meta-Learning via PAC-Bayes and Uniform Stability

Princeton University(普林斯顿大学)

Alec Farid, Anirudha Majumdar

摘要:

我们的动机是在元学习的背景下提供强泛化保证的问题。现有的泛化界限要么是具有挑战性的评估或提供空洞的保证,即使在相对简单的设置。我们推导出一个可能近似正确(PAC)的范围为基于梯度的元学习使用两种不同的泛化框架,以处理定性不同的挑战,在“基地”和“Meta”水平的泛化。我们采用统一稳定的算法在基础水平和PAC贝叶斯框架在Meta水平的界限。这种方法的结果是一种新的PAC界限,当基本学习者快速适应时,这种界限会更紧,这正是元学习的目标。我们表明,我们的界提供了一个更严格的保证比其他的单位球上的玩具非凸问题和基于文本的分类例子的界限。我们还提出了一个实际的正则化方案的动机在设置的边界是松散的,并证明了改进的性能基线技术。

英文摘要:

We are motivated by the problem of providing strong generalization guarantees in the context of meta-learning. Existing generalization bounds are either challenging to evaluate or provide vacuous guarantees in even relatively simple settings. We derive a probably approximately correct (PAC) bound for gradient-based meta-learning using two different generalization frameworks in order to deal with the qualitatively different challenges of generalization at the "base" and "meta" levels. We employ bounds for uniformly stable algorithms at the base level and bounds from the PAC-Bayes framework at the meta level. The result of this approach is a novel PAC bound that is tighter when the base learner adapts quickly, which is precisely the goal of meta-learning. We show that our bound provides a tighter guarantee than other bounds on a toy non-convex problem on the unit sphere and a text-based classification example. We also present a practical regularization scheme motivated by the bound in settings where the bound is loose and demonstrate improved performance over baseline techniques.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑