http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Extreme Learning을 통한 Convolutional Neural Network의 빠른 학습
유영우(Young-Woo Yoo),오세영(Se-Young Oh) 대한전자공학회 2015 대한전자공학회 학술대회 Vol.2015 No.11
CNN has been one of the best classifiers for images and object recognition. However, BP, mostly used for training CNN, takes a long time. To speed up training, a new architecture called CNN-ELM has been proposed here. It is based on a local image version of the ELM-AE learning. Using matlab 2015, our experiment shows a comparable classification performance to the BP trained CNN, with its training up to 200 times faster for MNIST and CIFAR-10 datasets.