Results 1 - 10 of 109

Playing Atari with Deep Reinforcement Learning

by Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Alex Graves, Ioannis Antonoglou, Daan Wierstra, Martin Riedmiller
"... We present the first deep learning model to successfully learn control policies di-rectly from high-dimensional sensory input using reinforcement learning. The model is a convolutional neural network, trained with a variant of Q-learning, whose input is raw pixels and whose output is a value functio ..."
Abstract - Cited by 31 (0 self) - Add to MetaCart
We present the first deep learning model to successfully learn control policies di-rectly from high-dimensional sensory input using reinforcement learning. The model is a convolutional neural network, trained with a variant of Q-learning, whose input is raw pixels and whose output is a value

Reinforcement Learning with Deep Architectures

by Daniel Selsam
"... There is both theoretical and empirical evidence that deep architectures may be more appropriate than shallow architectures for learning functions which exhibit hierarchical structure, and which can represent high level abstractions. An important development in machine learning research in the past ..."
Abstract - Add to MetaCart
few years has been a collection of algorithms that can train various deep architectures effectively. These methods have already led to many successes in the areas of supervised and unsupervised learning. They may prove to be just as useful in reinforcement learning as well, since solving a

Deep Reinforcement Learning in Keepaway Soccer

by Mateusz Kurek
"... This thesis focuses on deep neural networks and reinforcement learning (a.k.a deep reinforcement learning) in the context of the keepaway soccer problem. Deep learning is a new area in machine learning, but it already produced numer-ous successful applications. It is capable of learning high-level a ..."
Abstract - Add to MetaCart
This thesis focuses on deep neural networks and reinforcement learning (a.k.a deep reinforcement learning) in the context of the keepaway soccer problem. Deep learning is a new area in machine learning, but it already produced numer-ous successful applications. It is capable of learning high

Classifying Options for Deep Reinforcement Learning

by Kai Arulkumaran , Nat Dilokthanakul , Murray Shanahan , Anil Anthony Bharath
"... Abstract Deep reinforcement learning is the learning of multiple levels of hierarchical representations for reinforcement learning. Hierarchical reinforcement learning focuses on temporal abstractions in planning and learning, allowing temporally-extended actions to be transferred between tasks. In ..."
Abstract - Add to MetaCart
Abstract Deep reinforcement learning is the learning of multiple levels of hierarchical representations for reinforcement learning. Hierarchical reinforcement learning focuses on temporal abstractions in planning and learning, allowing temporally-extended actions to be transferred between tasks

Deep Auto-Encoder Neural Networks in Reinforcement Learning

by Sascha Lange, Martin Riedmiller
"... Abstract — This paper discusses the effectiveness of deep autoencoder neural networks in visual reinforcement learning (RL) tasks. We propose a framework for combining deep autoencoder neural networks (for learning compact feature spaces) with recently-proposed batch-mode RL algorithms (for learning ..."
Abstract - Cited by 16 (2 self) - Add to MetaCart
Abstract — This paper discusses the effectiveness of deep autoencoder neural networks in visual reinforcement learning (RL) tasks. We propose a framework for combining deep autoencoder neural networks (for learning compact feature spaces) with recently-proposed batch-mode RL algorithms (for

Efficient deep web crawling using reinforcement learning

by Lu Jiang , Zhaohui Wu , Qian Feng , Jun Liu , Qinghua Zheng - Advances in Knowledge Discovery and Data Mining , 2010
"... Abstract. Deep web refers to the hidden part of the Web that remains unavailable for standard Web crawlers. To obtain content of Deep Web is challenging and has been acknowledged as a significant gap in the coverage of search engines. To this end, the paper proposes a novel deep web crawling framew ..."
Abstract - Cited by 4 (2 self) - Add to MetaCart
framework based on reinforcement learning, in which the crawler is regarded as an agent and deep web database as the environment. The agent perceives its current state and selects an action (query) to submit to the environment according to Q-value. The framework not only enables crawlers to learn a

Sample-efficient Deep Reinforcement Learning for Dialog Control

by Kavosh Asadi , Jason D Williams
"... Abstract Representing a dialog policy as a recurrent neural network (RNN) is attractive because it handles partial observability, infers a latent representation of state, and can be optimized with supervised learning (SL) or reinforcement learning (RL). For RL, a policy gradient approach is natural ..."
Abstract - Add to MetaCart
Abstract Representing a dialog policy as a recurrent neural network (RNN) is attractive because it handles partial observability, infers a latent representation of state, and can be optimized with supervised learning (SL) or reinforcement learning (RL). For RL, a policy gradient approach

Language understanding for textbased games using deep reinforcement learning

by Karthik Narasimhan, Tejas D Kulkarni, Regina Barzilay - In Proceedings of the Conference on Empirical Methods in Natural Language Processing , 2015
"... In this paper, we consider the task of learn-ing control policies for text-based games. In these games, all interactions in the vir-tual world are through text and the un-derlying state is not observed. The re-sulting language barrier makes such envi-ronments challenging for automatic game players. ..."
Abstract - Cited by 4 (0 self) - Add to MetaCart
. We employ a deep reinforcement learning framework to jointly learn state representations and action policies using game rewards as feedback. This frame-work enables us to map text descriptions into vector representations that capture the semantics of the game states. We eval-uate our approach on two

Evolving deep unsupervised convolutional networks for vision-based reinforcement learning

by Jan Koutník, Jürgen Schmidhuber, Faustino Gomez - In Proceedings of the 2014 Genetic and Evolutionary Computation Conference (GECCO , 2014
"... Dealing with high-dimensional input spaces, like visual in-put, is a challenging task for reinforcement learning (RL). Neuroevolution (NE), used for continuous RL problems, has to either reduce the problem dimensionality by (1) compress-ing the representation of the neural network controllers or (2) ..."
Abstract - Cited by 4 (1 self) - Add to MetaCart
Dealing with high-dimensional input spaces, like visual in-put, is a challenging task for reinforcement learning (RL). Neuroevolution (NE), used for continuous RL problems, has to either reduce the problem dimensionality by (1) compress-ing the representation of the neural network controllers or (2

Faster Reinforcement Learning After Pretraining Deep Networks to Predict State Dynamics

by Charles W. Anderson, Minwoo Lee, Daniel L. Elliott
"... Abstract—Deep learning algorithms have recently appeared that pretrain hidden layers of neural networks in unsupervised ways, leading to state-of-the-art performance on large classifica-tion problems. These methods can also pretrain networks used for reinforcement learning. However, this ignores the ..."
Abstract - Add to MetaCart
Abstract—Deep learning algorithms have recently appeared that pretrain hidden layers of neural networks in unsupervised ways, leading to state-of-the-art performance on large classifica-tion problems. These methods can also pretrain networks used for reinforcement learning. However, this ignores
Results 1 - 10 of 109