Cooperative Initialization based Deep Neural Network Training

Abstract

Researchers have proposed various activation functions. These activation functions help the deep network to learn non-linear behavior with a significant effect on training dynamics and task performance. The performance of these activations also depends on the initial state of the weight parameters, i.e., different initial state leads to a difference in the performance of a network. In this paper, we have proposed a cooperative initialization for training the deep network using ReLU activation function to improve the network performance. Our approach uses multiple activation functions in the initial few epochs for the update of all sets of weight parameters while training the network. These activation functions cooperate to overcome their drawbacks in the update of weight parameters, which in effect learn better "feature representation" and boost the network performance later. Cooperative initialization based training also helps in reducing the overfitting problem and does not increase the number of parameters, inference (test) time in the final model while improving the performance. Experiments show that our approach outperforms various baselines and, at the same time, performs well over various tasks such as classification and detection. The Top-1 classification accuracy of the model trained using our approach improves by 2.8% for VGG-16 and 2.1% for ResNet-56 on CIFAR-100 dataset.

Learning Speaker-specific Lip-to-Speech Generation

“…Lip-reading also has crucial applications in helping speech-impaired people, such as hearing aids and generation of voice for people who communicate using lip movements only as they are unable to generate voice(aphonia). The advent of deep neural network models [11]- [15] and the availability of huge labeled datasets have helped in achieving a significant improvement in the task of the lip to text generation to solve a great variety of applications [6], [16]. These datasets consist of a large vocabulary in an unconstrained environment with thousands of speakers.…”