Efficient architecture for deep neural networks with heterogeneous sensitivity.

In this study, we present a neural network that consists of nodes with heterogeneous sensitivity. Each node in a network is assigned a variable that determines the sensitivity with which it learns to perform a given task. The network is trained via a constrained optimization that maximizes the sparsity of the sensitivity variables while ensuring optimal network performance. As a result, the network learns to perform a given task using only a few sensitive nodes. Insensitive nodes, which are nodes with zero sensitivity, can be removed from a trained network to obtain a computationally efficient network. Removing zero-sensitivity nodes has no effect on the performance of the network because the network has already been trained to perform the task without them. The regularization parameter used to solve the optimization problem was simultaneously found during the training of the networks. To validate our approach, we designed networks with computationally efficient architectures for various tasks such as autoregression, object recognition, facial expression recognition, and object detection using various datasets. In our experiments, the networks designed by our proposed method provided the same or higher performances but with far less computational complexity.

Deep neural networks with a set of node-wise varying activation functions.

In this study, we present deep neural networks with a set of node-wise varying activation functions. The feature-learning abilities of the nodes are affected by the selected activation functions, where the nodes with smaller indices become increasingly more sensitive during training. As a result, the features learned by the nodes are sorted by the node indices in order of their importance such that more sensitive nodes are related to more important features. The proposed networks learn input features but also the importance of the features. Nodes with lower importance in the proposed networks can be pruned to reduce the complexity of the networks, and the pruned networks can be retrained without incurring performance losses. We validated the feature-sorting property of the proposed method using both shallow and deep networks as well as deep networks transferred from existing networks.

Facial Attribute Recognition by Recurrent Learning With Visual Fixation.

This paper presents a recurrent learning-based facial attribute recognition method that mimics human observers' visual fixation. The concentrated views of a human observer while focusing and exploring parts of a facial image over time are generated and fed into a recurrent network. The network makes a decision concerning facial attributes based on the features gleaned from the observer's visual fixations. Experiments on facial expression, gender, and age datasets show that applying visual fixation to recurrent networks improves recognition rates significantly. The proposed method not only outperforms state-of-the-art recognition methods based on static facial features, but also those based on dynamic facial features.