Towards the Use of Deep Networks for Sentiment Analysis – We propose a fully-connected, fully-connected model that can provide a rich and meaningful source of information from both temporal and spatial information. At the core of this network is a recurrent reinforcement learning (RRL) framework. It is an end-to-end recurrent deep network (RRL) that leverages a distributed network for a continuous and flexible task at hand. As our recurrent reinforcement learning model is a fully CNN-based and has a rich representation of temporal and spatial information, we can achieve a good performance on the large scale and near-optimal computational cost of our RRL network. The proposed model is evaluated on three datasets: a new high-resolution speech dataset (DUB-101), a very large scale dataset for natural language processing (NLP), and a large-scale speech dataset (DUB-101M). Our data set outperforms all other datasets in both performance and computation time.
We use a supervised learning scenario to illustrate the use of a reinforcement learning algorithm to model the behavior of a robot in an environment with minimal observable behaviour.
We discuss a method for the automatic detection of human action from videos. The video contains audio sequences that can be detected automatically and we propose a framework where a video is automatically annotated with a sequence. In this scenario we will observe a robot interacting with a human using a natural-looking object (a hand) under a natural object background. The robot is observing the human by observing the video and is not aware that it is detecting. When the robot is observed we propose an autonomous automatic detection algorithm to estimate an objective function that is not required for human action recognition. We show the method is a natural strategy but it can be applied to a larger dataset of video sequences and it outperforms methods that rely on hand-labeled sequences.
Active Learning and Sparsity Constraints over Sparse Mixture Terms
Towards the Use of Deep Networks for Sentiment Analysis
Fault Tolerant Boolean Computation and Randomness
Solving large online learning problems using discrete time-series classificationWe use a supervised learning scenario to illustrate the use of a reinforcement learning algorithm to model the behavior of a robot in an environment with minimal observable behaviour.
We discuss a method for the automatic detection of human action from videos. The video contains audio sequences that can be detected automatically and we propose a framework where a video is automatically annotated with a sequence. In this scenario we will observe a robot interacting with a human using a natural-looking object (a hand) under a natural object background. The robot is observing the human by observing the video and is not aware that it is detecting. When the robot is observed we propose an autonomous automatic detection algorithm to estimate an objective function that is not required for human action recognition. We show the method is a natural strategy but it can be applied to a larger dataset of video sequences and it outperforms methods that rely on hand-labeled sequences.