W
Wonyong Sung
Researcher at Seoul National University
Publications - 222
Citations - 7039
Wonyong Sung is an academic researcher from Seoul National University. The author has contributed to research in topics: Recurrent neural network & SIMD. The author has an hindex of 34, co-authored 218 publications receiving 6373 citations. Previous affiliations of Wonyong Sung include Gwangju Institute of Science and Technology.
Papers
More filters
Journal ArticleDOI
A statistical model-based voice activity detection
TL;DR: An effective hang-over scheme which considers the previous observations by a first-order Markov process modeling of speech occurrences is proposed which shows significantly better performances than the G.729B VAD in low signal-to-noise ratio (SNR) and vehicular noise environments.
Journal ArticleDOI
Structured Pruning of Deep Convolutional Neural Networks
TL;DR: The proposed work shows that when pruning granularities are applied in combination, the CIFAR-10 network can be pruned by more than 70% with less than a 1% loss in accuracy.
Posted Content
Structured Pruning of Deep Convolutional Neural Networks
TL;DR: In this article, the importance weight of each particle is assigned by computing the misclassification rate with corresponding connectivity pattern, and the pruned network is re-trained to compensate for the losses due to pruning.
Proceedings ArticleDOI
Fixed-point feedforward deep neural network design using weights +1, 0, and −1
Kyuyeon Hwang,Wonyong Sung +1 more
TL;DR: The designed fixed-point networks with ternary weights (+1, 0, and -1) and 3-bit signal show only negligible performance loss when compared to the floating-point coun-terparts.
Proceedings ArticleDOI
Fixed point optimization of deep convolutional neural networks for object recognition
TL;DR: The results indicate that quantization induces sparsity in the network which reduces the effective number of network parameters and improves generalization, and reduces the required memory storage by a factor of 1/10 and achieves better classification results than the high precision networks.