Neural Embeddings for Sentiment Classification – We present an innovative approach for embedding Chinese and English texts in a unified framework, which is able to process and embed sentences in an end-to-end fashion.
We present a unified framework for automatic feature learning systems. In particular, it is proposed to utilize features from text-to-text to predict a human-level representation of a word in the English language. The proposed approach is based on a recurrent neural network (RNN) that is trained with a novel recurrent language model that represents word vectors over a sequence of text-structured data. Then, the RNN-RNN encodes the data using a convolutional neural network to perform word prediction. Our experiments on a dataset of English sentences show that the proposed method is effective and comparable to RNN-RNN’s performance for predicting speech words in Chinese text. We also describe a new approach for automatic feature learning in sentence embeddings.
The task of learning to see people in an immersive game requires the player to make decisions and manipulate their environment. The choice of player viewpoint is crucial in a large variety of human, virtual, cognitive and collaborative games. In the long term, we aim to learn to see people by learning a new visual feature that is useful for the user to manipulate with the ability to navigate around virtual spaces. We present a multi-view model, which is adapted to the user’s choice in the first place, and use its knowledge to represent a user’s own vision. It can use objects and objects from both their human perspective, and objects and objects from the user’s own vision. We achieve an improvement of 13.8% on average over the baseline state-of-the-art, with a mean top-1 accuracy of 83.13%.
Object Category Recognition in Video by Joint Stereo Matching and Sparse Coding
A Bayesian Approach to Learn with Sparsity-Contrastive Multiplicative Task-Driven Data
Neural Embeddings for Sentiment Classification
Symbolism and Cognition in a Neuronal Perceptron
DeepFace: Learning to see people in real-timeThe task of learning to see people in an immersive game requires the player to make decisions and manipulate their environment. The choice of player viewpoint is crucial in a large variety of human, virtual, cognitive and collaborative games. In the long term, we aim to learn to see people by learning a new visual feature that is useful for the user to manipulate with the ability to navigate around virtual spaces. We present a multi-view model, which is adapted to the user’s choice in the first place, and use its knowledge to represent a user’s own vision. It can use objects and objects from both their human perspective, and objects and objects from the user’s own vision. We achieve an improvement of 13.8% on average over the baseline state-of-the-art, with a mean top-1 accuracy of 83.13%.