Adaptive Gradient-Based Meta-Learning Methods
Adaptive Gradient-Based Meta-Learning Methods
复制标题
DOI：
--
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
M. Khodak;Maria-Florina Balcan;Ameet Talwalkar
中科院分区：
文献类型：
--
作者：
M. Khodak;Maria-Florina Balcan;Ameet Talwalkar
我们建立了一个理论框架，用于设计和理解实用的元学习方法，该方法将任务相似性的复杂形式化与在线凸优化和顺序预测算法的广泛文献相结合。我们的方法使任务相似性自适应学习，提供更清晰的转移风险界限的设置统计学习学习，并导致直接推导出平均情况下的遗憾界限的有效算法的设置中的任务环境动态变化或任务共享一定的几何结构。我们使用我们的理论来修改几个流行的元学习算法，并提高他们的元测试时间性能的标准问题，在几次学习和联邦学习。
We build a theoretical framework for designing and understanding practical meta-learning methods that integrates sophisticated formalizations of task-similarity with the extensive literature on online convex optimization and sequential prediction algorithms. Our approach enables the task-similarity to be learned adaptively, provides sharper transfer-risk bounds in the setting of statistical learning-to-learn, and leads to straightforward derivations of average-case regret bounds for efficient algorithms in settings where the task-environment changes dynamically or the tasks share a certain geometric structure. We use our theory to modify several popular meta-learning algorithms and improve their meta-test-time performance on standard problems in few-shot learning and federated learning.