DC Field | Value | Language |
---|---|---|
dc.contributor.author | 이윤호 | - |
dc.date.accessioned | 2018-10-17T05:48:36Z | - |
dc.date.available | 2018-10-17T05:48:36Z | - |
dc.date.issued | 2018 | - |
dc.identifier.other | OAK-2015-08132 | - |
dc.identifier.uri | http://postech.dcollection.net/common/orgView/200000104867 | ko_KR |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/93592 | - |
dc.description | Master | - |
dc.description.abstract | Deep learning has been tremendously successful in many difficult tasks including image classification and game-playing. However, deep networks require copious amounts of data in order to achieve such performance. Meta-learning methods have recently gained attention in this context as a way to remedy this dependence on large datasets. Gradient-based meta-learning methods leverage gradient descent to learn the commonalities among various tasks. While previous such methods have been successful in meta-learning tasks, they resort to simple gradient descent during meta-testing. Our primary contribution is the MT-net , which enables the meta- learner to learn on each layer’s activation space a subspace that the task-specific learner performs gradient descent on. Additionally, a task-specific learner of an MT-net performs gradient descent with respect to a meta-learned distance metric, which warps the activation space to be more sensitive to task identity. We demonstrate that the dimension of this learned subspace reflects the complexity of the task-specific learner’s adaptation task, and also that our model is less sensitive to the choice of initial learning rates than previous gradient-based meta-learning methods. Our method achieves state-of-the-art or comparable performance on few-shot classification and regression tasks. | - |
dc.language | eng | - |
dc.publisher | 포항공과대학교 | - |
dc.title | Gradient-based Meta-learning with Learned Layerwise Metric and Subspace | - |
dc.title.alternative | 학습된 거리함수와 부분공간을 이용한 경사도 기반 메타러닝 | - |
dc.type | Thesis | - |
dc.contributor.college | 일반대학원 컴퓨터공학과 | - |
dc.date.degree | 2018- 8 | - |
dc.type.docType | Thesis | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.