Learning Sparse Representations of Data with Regularized Dropout – The approach is based on the idea that if a data-driven model is designed to capture the information in the real world, then it must be able to capture and interpret this information. However, this is rarely considered. This paper presents an in-depth analysis into the learning of a well-adapted deep learning model, namely the convolutional neural network (CNN)-CNF, and the use of such a model for machine learning problems. To our best knowledge, this is the first research into this framework, with the main importance being to show that as a prerequisite, the CNN has to learn to learn the information from a data-driven architecture. The experimental results show that our approach is able to outperform standard CNNs with significant improvement on two datasets, namely the recently developed IJB-2D dataset and the popular SVHN dataset. The CNN-CNF is particularly good for the IJB dataset, achieving state-of-the-art performance on both datasets, with some limitations.

We present an adaptive sparse coding of neural networks to classify complex objects. With adaptive sparse coding, neurons in the input layer are connected to the global network of synaptic weights. In this way, if the network can be modelled on a given model, an adaptive coding system can be developed, based on such a network. We show that this adaptive coding scheme is more efficient than the model-based one by approximately solving the problem of learning sparse coding in a non-linear fashion. In particular, for an adaptive sparse coding system, an adaptive coding neural network can be trained using recurrent neural networks, without using any prior information on the current model.

Poseidon: An Efficient Convolutional Neural Network for Automatic Detection of Severe Sleep Apnea

Optimal Bounds for Online Convex Optimization Problems via Random Projections

# Learning Sparse Representations of Data with Regularized Dropout

Learning the Block Kernel for Sparse Subspace Analysis with Naive Bayes

The Multi-Domain VisionNet: A Large-scale 3D Wide-RoboDetector Dataset for Pathological Lung Nodule DetectionWe present an adaptive sparse coding of neural networks to classify complex objects. With adaptive sparse coding, neurons in the input layer are connected to the global network of synaptic weights. In this way, if the network can be modelled on a given model, an adaptive coding system can be developed, based on such a network. We show that this adaptive coding scheme is more efficient than the model-based one by approximately solving the problem of learning sparse coding in a non-linear fashion. In particular, for an adaptive sparse coding system, an adaptive coding neural network can be trained using recurrent neural networks, without using any prior information on the current model.