Show and Tell: Learning to Watch from Text Videos – We present an interactive text-to-speech system, SentientReSPRESS, that automatically determines what’s being shown by a text. SentientReSPRESS is a natural language processing system, which integrates deep learning with machine learning; and, in a more practical way, it is inspired by the well-known machine learning paradigm: the neural network. SentientReSPRESS features multiple state space of text and sentences; it learns the sentence structure from input sentences. SentientReSPRESS utilizes convolutional neural network architecture to learn the sentence structure. SentientReSPRESS is trained on a corpus of 30K sentences, and then tested to find the best sentence structure by analyzing the sentence similarity to the source data for our task. SentientReSPRESS has learned over 8K sentences, while using 2.8M parameters.

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.

Learning to Acquire Information from Noisy Speech

A General Method for Scalable Convex Optimization

Show and Tell: Learning to Watch from Text Videos

Predictive Nonlinearity in Linear-Quadratic Control Problems

Learning Sparse Representations of Data with Regularized DropoutThe 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.