DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Hwang, Sung Ju | - |
dc.contributor.advisor | 황성주 | - |
dc.contributor.author | Park, Geon | - |
dc.date.accessioned | 2023-06-22T19:31:30Z | - |
dc.date.available | 2023-06-22T19:31:30Z | - |
dc.date.issued | 2023 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1032321&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/308234 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 김재철AI대학원, 2023.2,[iv, 36 p. :] | - |
dc.description.abstract | Neural Network Quantization aims to reduce the size and computational complexity of a neural network for more efficient training and inference of neural networks. However, existing methods often render themselves impractical in real-world scenarios, such as On-device Federated Learning, and with compact models such as MobileNet. In this paper, We show that applying Neural Network quantization in these scenarios are difficult. In On-device Federated Learning scenarios, many diverse devices with different hardware constraints can participate in the same Federated Learning, which leads to degenerate performance in the high-performance devices. With compact models with less redundancies in the weights, it is much more difficult to find quantized weights that do not incur a drop in the model's accuracy. We tackle these challenges by introducing two novel methods for practical neural network quantization: Bitwidth Heterogeneous Federated Learning with Progressive Weight Dequantizer, and Neural Network Binarization with Task-dependent Aggregated Transform. We show that these methods are effective at applying Neural Network quantization in the aforementioned practical scenarios. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Neural network quantization▼aEfficient neural network training▼aEfficient neural network inference▼aFederated learning▼aBitwidth-heterogeneous federated learning | - |
dc.subject | 인공신경망 양자화▼a효율적인 인공신경망 학습▼a효율적인 인공신경망 추론▼a연합학습▼a이종 정밀도 연합학습 | - |
dc.title | Quantized neural network training and inference in practical scenarios | - |
dc.title.alternative | 실용적인 시나리오에서의 인공신경망 학습과 추론의 양자화 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :김재철AI대학원, | - |
dc.contributor.alternativeauthor | 박건 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.