Is Bayesian Model-Agnostic Meta Learning Better than Model-Agnostic Meta Learning, Provably?
Is Bayesian Model-Agnostic Meta Learning Better than Model-Agnostic Meta Learning, Provably?
复制标题
贝叶斯模型无关元学习是否比模型无关元学习更好?
DOI:
10.48550/arxiv.2203.03059
复制
发表时间:
2022
期刊:
影响因子:
--
通讯作者:
Tianyi
中科院分区:
文献类型:
--
作者:
Lisha Chen;Tianyi
Meta learning aims at learning a model that can quickly adapt to unseen tasks. Widely used meta learning methods include model agnostic meta learning (MAML), implicit MAML, Bayesian MAML. Thanks to its ability of modeling uncertainty, Bayesian MAML often has advantageous empirical performance. However, the theoretical understanding of Bayesian MAML is still limited, especially on questions such as if and when Bayesian MAML has provably better performance than MAML. In this paper, we aim to provide theoretical justifications for Bayesian MAML's advantageous performance by comparing the meta test risks of MAML and Bayesian MAML. In the meta linear regression, under both the distribution agnostic and linear centroid cases, we have established that Bayesian MAML indeed has provably lower meta test risks than MAML. We verify our theoretical results through experiments.
登录
查看更多内容
DOI:
--
发表时间:
2018
期刊:
--
影响因子:
--
作者:
Giulia Denevi;C. Ciliberto;Dimitris Stamos;M. Pontil
通讯作者:
Giulia Denevi;C. Ciliberto;Dimitris Stamos;M. Pontil
DOI:
--
发表时间:
2019-09
期刊:
--
影响因子:
--
作者:
A. Rajeswaran;Chelsea Finn;S. Kakade;S. Levine
通讯作者:
A. Rajeswaran;Chelsea Finn;S. Kakade;S. Levine
DOI:
--
发表时间:
2020-10
期刊:
ArXiv
影响因子:
--
作者:
Yu Bai;Minshuo Chen;Pan Zhou;T. Zhao;J. Lee;S. Kakade;Haiquan Wang;Caiming Xiong
通讯作者:
Yu Bai;Minshuo Chen;Pan Zhou;T. Zhao;J. Lee;S. Kakade;Haiquan Wang;Caiming Xiong
影响因子:
6
作者:
Song, Jiang-Ling;Hu, Wenfeng;Zhang, Rui
通讯作者:
Zhang, Rui