期刊
IEEE TRANSACTIONS ON CYBERNETICS
卷 53, 期 6, 页码 3440-3453出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCYB.2021.3123403
关键词
Task analysis; Decoding; Training; Testing; Measurement; Optimization; Mathematical models; Decoder; ensemble learning; latent variable; metalearning
This article presents a novel metalearning method that controls the gradient descent process in a neural network by limiting the model parameters in a low-dimensional latent space. It also introduces an alternative design of the decoder with shared weights to reduce the number of parameters. Experimental results show that the proposed approach outperforms the state of the art in classification tasks.
Metalearning has been widely applied for implementing few-shot learning and fast model adaptation. Particularly, existing metalearning methods have been exploited to learn the control mechanism for gradient descent processes, in an effort to facilitate gradient-based learning in gaining high speed and generalization ability. This article presents a novel method that controls the gradient descent process of the model parameters in a neural network, by limiting the model parameters within a low-dimensional latent space. The main challenge for implementing this idea is that a decoder with many parameters may be required. To tackle this problem, the article provides an alternative design of the decoder with a structure that shares certain weights, thereby reducing the number of required parameters. In addition, this work combines ensemble learning with the proposed approach to improve the overall learning performance. Systematic experimental studies demonstrate that the proposed approach offers results superior to the state of the art in performing the Omniglot classification and miniImageNet classification tasks.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据