Alpha MAML：自适应模型无关元学习

需积分: 24 127 浏览量更新于2024-09-01 2 收藏 883KB PDF 举报

"Alpha MAML: Adaptive Model-Agnostic Meta-Learning" 元学习（Meta-Learning）是一种机器学习方法，旨在使模型能够快速适应新任务，尤其在只有少量样本的情况下。MAML（Model-Agnostic Meta-Learning，模型非特定元学习）是元学习的一种著名算法，它通过在多种学习任务上训练模型，使其能迅速适应新的任务。MAML的核心思想是通过优化模型的初始参数，使得模型经过少数几次迭代就能达到较好的性能。然而，MAML的一个主要缺点是需要大量的超参数调整以保证训练过程的稳定。这不仅增加了计算成本，也使得应用MAML变得困难。针对这一问题，研究者提出了Alpha MAML，一个MAML的扩展版本，它融入了在线超参数自适应方案，可以自动调整超参数，从而消除对元学习率和学习率手动调整的需求。 Alpha MAML的工作原理是在训练过程中动态地适应元学习率和学习率，这使得模型能够更好地自我调整，提高训练稳定性，并减少对超参数敏感性的依赖。在6th ICML Workshop on Automated Machine Learning (2019)上发表的论文中，作者Harkirat Singh Behl、Atılım Gunes Baydin和Philip H.S. Torr展示了Alpha MAML在Omniglot数据集上的应用，结果显示，Alpha MAML显著减少了对MAML训练超参数的调整需求，同时提高了训练稳定性，降低了对超参数选择的敏感性。元学习在分类、回归以及强化学习中的策略梯度微调等领域都展现出了良好的性能。Alpha MAML的出现，使得元学习方法在实际应用中更加便捷，对于快速适应不断变化的任务环境具有重要意义，特别是在数据稀疏或实时学习的场景下。总结来说，Alpha MAML是MAML算法的一种优化，解决了原有MAML方法中需要大量超参数调整的问题，通过引入在线超参数自适应机制，提高了模型的训练效率和泛化能力，降低了对人工调整的依赖，为元学习的实际应用提供了更优的解决方案。这一方法对于未来元学习领域的研究和发展具有重要价值，特别是在自动化机器学习领域，它可以进一步推动模型的自适应能力和学习效率。

6th ICML Workshop on Automated Machine Learning (2019)

Alpha MAML: Adaptive Model-Agnostic Meta-Learning

Harkirat Singh Behl harkirat@robots.ox.ac.uk

Atılım G¨une¸s Baydin gunes@robots.ox.ac.uk

Philip H.S. Torr phst@robots.ox.ac.uk

University of Oxford

Abstract

Model-agnostic meta-learning (MAML) is a meta-learning technique to train a model on a

multitude of learning tasks in a way that primes the model for few-shot learning of new

tasks. The MAML algorithm performs well on few-shot learning problems in classiﬁcation,

regression, and ﬁne-tuning of policy gradients in reinforcement learning, but comes with the

need for costly hyperparameter tuning for training stability. We address this shortcoming

by introducing an extension to MAML, called Alpha MAML, to incorporate an online

hyperparameter adaptation scheme that eliminates the need to tune meta-learning and

learning rates. Our results with the Omniglot database demonstrate a substantial reduction

in the need to tune MAML training hyperparameters and improvement to training stability

with less sensitivity to hyperparameter choice.

1. Introduction

Meta-learning—or “learning to learn”—concerns machine learning models that can improve

their learning quality by altering aspects of the learning process such as the model architecture,

optimization rules, initialization, or learning hyperparameters (Thrun and Pratt, 2012;

Schmidhuber, 1987; Hochreiter et al., 2001). An important application of meta-learning is in

few-shot learning problems (Vinyals et al., 2016; Behl et al., 2018), where one is concerned

with developing methods able to learn new concepts from one or only a few instances (Lake

et al., 2015). In this paper we focus on the state-of-the-art model-agnostic meta-learning

(MAML) (Finn et al., 2017) method, which is a conceptually simple and general algorithm

that has been shown to outperform existing approaches in tasks including few-shot image

classiﬁcation and few-shot adaptation in reinforcement learning (Antoniou et al., 2019).

MAML aims to solve the few-shot learning problem by being just few gradient descent

steps away from any new concepts, doing so by making the assumption that learning a new

concept will just involve few parameter updates (Algorithm 1). In other words, MAML is

based on learning an initial representation that can be eﬃciently ﬁne-tuned for new tasks in

a few steps.

The generality of MAML comes with the diﬃculty of choosing hyperparameters to

achieve stable training in practice (Antoniou et al., 2019). MAML has two important

hyper-parameters, namely the learning rate

and the meta-learning rate

, thus increasing

any hyperparameter grid search computation by an order, and making it signiﬁcantly more

time and resource consuming than comparable methods. Another complication to this

problem is the fact that it is currently not established whether the technique can beneﬁt

from a conventional decaying schedule for the inner learning rate

. Furthermore, a good

value of α in MAML is even more important than for any conventional stochastic gradient

descent (SGD) optimization, because only a handful of samples are available in the few-shot

learning case. This has signiﬁcant consequences, making it diﬃcult to scale this algorithm

2019 H.S. Behl, A.G. Baydin, and P.H.S. Torr.

arXiv:1905.07435v1 [cs.LG] 17 May 2019

下载后可阅读完整内容，剩余9页未读，立即下载

zyp18810010105

粉丝: 0
资源: 6

Alpha MAML：自适应模型无关元学习

元学习-MAML-资源整合

matlab基于改进模型不可知元学习算法的快速自适应主动噪声控制.zip

PMLSM伺服系统的特征模型与自适应迭代学习控制

请参阅用于无监督多目标域自适应的激光雷达不可知3D检测框架_See Eye to Eye A Lidar-Agnostic 3D

基于灰色马尔科夫模型的采煤机自适应截割策略研究

参数不确定时滞系统的自适应广义函数投影滞后同步

多智能体系统自适应跟踪控制.pdf

基于自适应卡尔曼滤波原理的车距预测研究

基于Petri网的自主网络服务自适应兼容性分析

无需速度测量的网络机器人系统的自适应同步控制

最新资源