DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kim, Jun Mo | - |
dc.contributor.advisor | 김준모 | - |
dc.contributor.author | Lee, Jang Hyeon | - |
dc.date.accessioned | 2018-06-20T06:23:06Z | - |
dc.date.available | 2018-06-20T06:23:06Z | - |
dc.date.issued | 2017 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=718703&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/243370 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2017.8,[iii, 14 p. :] | - |
dc.description.abstract | Deep learning with artificial neural networks is one of the most powerful AI technologies. It shows state-of-the-art performance in various fields. To save network training time and improve performance, transfer learning is commonly used. Transfer learning achieves good target performance, but source performance is poor due to catastrophic forgetting. We propose a novel continual learning method called Reversible and Mergeable Learning (RML). The RML method keeps all the original parameters to enable the reversible process of additional training on the target task. After training, the original network and newly learned network can be merged into a network of the same size with the original one. A new network trained with the RML method shows better performance than the finetuning method, and surprisingly, can even outperform the original one on the source task without seeing any source data when the source and target tasks are similar. The RML method can also be used for data incremental learning. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Continual Learning▼aMultitask Learning▼aData Incremental Learning▼aArtificial Neural Network▼aDeep Learning | - |
dc.subject | 지속 학습▼a멀티 태스킹 학습▼a데이터 증가 학습▼a인공 신경망▼a심층 학습 | - |
dc.title | Reversible and mergeable learning for continual learning of artificial neural networks | - |
dc.title.alternative | 인공 신경망의 지속 학습을 위한 가역 및 가합 학습 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전기및전자공학부, | - |
dc.contributor.alternativeauthor | 이장현 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.