DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 이문용 | - |
dc.contributor.author | Lee, Sol | - |
dc.contributor.author | 이솔 | - |
dc.date.accessioned | 2024-07-30T19:30:52Z | - |
dc.date.available | 2024-07-30T19:30:52Z | - |
dc.date.issued | 2024 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1096209&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/321424 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 데이터사이언스대학원, 2024.2,[iv, 39 p. :] | - |
dc.description.abstract | Sample selection is an effective method for robust learning in the presence of label noises. However, existing approaches that rely on small loss values to identify clean samples can commit the error of excluding clean samples with large losses, called hard clean samples. These hard clean samples play a crucial role in shaping high-quality decision boundaries and excluding them can lead to degraded generalization performance. Toward overcoming these limitations, this paper introduces a novel sample selection strategy called KALM, which utilizes an iterative and powerful model generation and filtering strategy based on softmax probabilities and loss values obtained from deep neural network outputs. KALM preserves challenging and correct-labeled samples while effectively removing label noises, contributing to the construction of a high-performing classifier. Notably, KALM does not rely on expensive prior information such as noise rates or clean validation data, and it produces robust performance across various noise types and ratios. Experimental results on CIFAR-10, CIFAR-100, and Clothing1M datasets consistently highlight the superior performance achieved by KALM compared to existing approaches. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | 강건한 심층 학습▼a지도 학습▼a분류▼a노이즈 라벨 | - |
dc.subject | Robust deep learning▼aSupervised learning▼aClassification▼aNoisy label | - |
dc.title | Preserving hard clean samples for robust learning with noisy labels in deep neural networks | - |
dc.title.alternative | 어렵고 올바른 라벨 샘플 보존을 통한 라벨 노이즈에 강건한 심층 신경망 학습법 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :데이터사이언스대학원, | - |
dc.contributor.alternativeauthor | Yi, Mun Yong | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.