DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Choo, Jaegul | - |
dc.contributor.advisor | 주재걸 | - |
dc.contributor.author | Lee, Sanghyeon | - |
dc.date.accessioned | 2022-04-15T07:56:31Z | - |
dc.date.available | 2022-04-15T07:56:31Z | - |
dc.date.issued | 2021 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=963743&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/294845 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : AI대학원, 2021.8,[iii, 12 p. :] | - |
dc.description.abstract | Graph Convolutional Networks(GCNs) are successful models for learning graph structures and representations. However, real-world graphs are extremely large with numerous nodes and edges. As computational complexity increases exponentially with the depth of the GCNs layer, learning such real-world graphs are expensive. To mitigate this, previous works propose sampling methods that utilize few nodes to aggregate neighbor information. These graph sampling methods assume that the sampled nodes are independent, but this is an incorrect assumption for real-world graphs, as they are heavily correlated. To address this problem, we propose a sequential sampling method inspired by Monte Carlo - Markov Chain algorithm. We conduct the experiments on graph benchmark dataset: Cora, Citeseer and Pumbed. Experimental results show that our proposed algorithm outperforms the previous sampling methods. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Convolutional Graph Network▼aGraph Sampling▼aMachine Learning▼aMarkov Chain - Monte Carlo▼aArtificial Intelligence | - |
dc.subject | 컨볼루션 그래프 인공 신경망▼a그래프 샘플링▼a기계학습▼a마르코프 체인 몬테 카를로▼a인공지능 | - |
dc.title | Sequential sampling for correlated graph sampling | - |
dc.title.alternative | 상관관계가 있는 그래프 샘플링을 위한 순차 샘플링 방법론 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :AI대학원, | - |
dc.contributor.alternativeauthor | 이상현 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.