Structural Matching through Reinforcement Learning – This paper addresses the problem of supervised learning of visual attention networks by applying deep reinforcement learning (DL) to reinforcement learning tasks. DL is an end-to-end learning algorithm that does not require the user to learn any specific visual scene. In particular, DL can learn to capture visual dependencies and to adapt to different visual cues of the scene at different levels of its complexity, in a global way. In this paper, we propose a novel DL model trained with the task-dependent visual cue to learn to predict the next action sequence over the entire network. As an example, we consider our attention-to-sequence learning algorithm which is trained from scratch and learns to predict the next sequence over every visual cue of an object at each level of the network (i.e. after training the supervised models only on the task-dependent visual cue). We demonstrate that our DL model outperforms the state-of-the-art attention based vision models in terms of accuracy, on an unstructured object detection task.
Deep learning has emerged as an important technology in medical applications, providing the tools to solve complex and frequently-constrained clinical tasks in medical systems. We show that deep neural networks can be used to learn the semantic meaning of concepts, and that such representations can be used to guide the user to help with medical decisions. We demonstrate how recurrent networks can be used to model concept representations and how representations can be learned from the training data using Convolutional Neural Networks. We evaluate these models on the challenging clinical domains, and compare them to state-of-the-art approaches including supervised learning, reinforcement learning, and deep learning-based approaches.
The Effectiveness of Multitask Learning in Deep Learning Architectures
Towards a theory of universal agents
Structural Matching through Reinforcement Learning
On the Generalizability of the Population Genetics Dataset
Efficient Large-scale Prediction of Time Series of Diabetic Retinopathy Patients Using Multi-Task LearningDeep learning has emerged as an important technology in medical applications, providing the tools to solve complex and frequently-constrained clinical tasks in medical systems. We show that deep neural networks can be used to learn the semantic meaning of concepts, and that such representations can be used to guide the user to help with medical decisions. We demonstrate how recurrent networks can be used to model concept representations and how representations can be learned from the training data using Convolutional Neural Networks. We evaluate these models on the challenging clinical domains, and compare them to state-of-the-art approaches including supervised learning, reinforcement learning, and deep learning-based approaches.