DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kim, Junmo | - |
dc.contributor.advisor | 김준모 | - |
dc.contributor.author | Lee, Jong Ho | - |
dc.date.accessioned | 2019-09-04T02:40:58Z | - |
dc.date.available | 2019-09-04T02:40:58Z | - |
dc.date.issued | 2018 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=733961&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/266746 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2018.2,[iv, 26 p. :] | - |
dc.description.abstract | In this thesis, we have proposed a new training method which adds auxiliary tasks to existing models. For visual question answering problem, it is important to increase the mutual information among questions, images, and answers. By reconstructing features of the training data from answer, we were able to guide learning process more efficiently. The proposed method is not limited to a specific model, and it can also improve the performance of model while preserving the size of models. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Deep Learning▼aVisual Question Answering▼aMulti-task Learning▼aImage Understanding | - |
dc.subject | 심층 학습▼a시각적 질의 응답▼a다 과제 학습▼a영상 이해 | - |
dc.title | Deep neural network with multimodal auxiliary tasks for visual question answering | - |
dc.title.alternative | 시각적 질의 응답 문제 해결을 위한 다양한 보조 과제를 이용한 신경망에 대한 연구 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전기및전자공학부, | - |
dc.contributor.alternativeauthor | 이종호 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.