TY - JOUR
T1 - Recent advances in convolutional neural networks
AU - Gu, Jiuxiang
AU - Wang, Zhenhua
AU - Kuen, Jason
AU - Ma, Lianyang
AU - Shahroudy, Amir
AU - Shuai, Bing
AU - Liu, Ting
AU - Wang, Xingxing
AU - Wang, Gang
AU - Cai, Jianfei
AU - Chen, Tsuhan
PY - 2018/5
Y1 - 2018/5
N2 - In the last few years, deep learning has led to very good performance on a variety of problems, such as visual recognition, speech recognition and natural language processing. Among different types of deep neural networks, convolutional neural networks have been most extensively studied. Leveraging on the rapid growth in the amount of the annotated data and the great improvements in the strengths of graphics processor units, the research on convolutional neural networks has been emerged swiftly and achieved state-of-the-art results on various tasks. In this paper, we provide a broad survey of the recent advances in convolutional neural networks. We detailize the improvements of CNN on different aspects, including layer design, activation function, loss function, regularization, optimization and fast computation. Besides, we also introduce various applications of convolutional neural networks in computer vision, speech and natural language processing.
AB - In the last few years, deep learning has led to very good performance on a variety of problems, such as visual recognition, speech recognition and natural language processing. Among different types of deep neural networks, convolutional neural networks have been most extensively studied. Leveraging on the rapid growth in the amount of the annotated data and the great improvements in the strengths of graphics processor units, the research on convolutional neural networks has been emerged swiftly and achieved state-of-the-art results on various tasks. In this paper, we provide a broad survey of the recent advances in convolutional neural networks. We detailize the improvements of CNN on different aspects, including layer design, activation function, loss function, regularization, optimization and fast computation. Besides, we also introduce various applications of convolutional neural networks in computer vision, speech and natural language processing.
KW - Convolutional neural network
KW - Deep learning
UR - http://www.scopus.com/inward/record.url?scp=85031710709&partnerID=8YFLogxK
U2 - 10.1016/j.patcog.2017.10.013
DO - 10.1016/j.patcog.2017.10.013
M3 - Article
AN - SCOPUS:85031710709
SN - 0031-3203
VL - 77
SP - 354
EP - 377
JO - Pattern Recognition
JF - Pattern Recognition
ER -