Human-level control through deep reinforcement learning

@article{Mnih2015HumanlevelCT,
  title={Human-level control through deep reinforcement learning},
  author={Volodymyr Mnih and Koray Kavukcuoglu and David Silver and Andrei A. Rusu and Joel Veness and Marc G. Bellemare and Alex Graves and Martin A. Riedmiller and Andreas Fidjeland and Georg Ostrovski and Stig Petersen and Charlie Beattie and Amir Sadik and Ioannis Antonoglou and Helen King and Dharshan Kumaran and Daan Wierstra and Shane Legg and Demis Hassabis},
  journal={Nature},
  year={2015},
  volume={518},
  pages={529-533}
}
The theory of reinforcement learning provides a normative account, deeply rooted in psychological and neuroscientific perspectives on animal behaviour, of how agents may optimize their control of an environment. To use reinforcement learning successfully in situations approaching real-world complexity, however, agents are confronted with a difficult task: they must derive efficient representations of the environment from high-dimensional sensory inputs, and use these to generalize past… Expand
Learning sparse representations in reinforcement learning
TLDR
It is hypothesize that such failures do not arise in the brain because of the ubiquitous presence of lateral inhibition in the cortex, producing sparse distributed internal representations that support the learning of expected future reward. Expand
On the Reliability and Generalizability of Brain-inspired Reinforcement Learning Algorithms
TLDR
This is the first attempt to formally test the possibility that computational models mimicking the way the brain solves general problems can lead to practical solutions to key challenges in machine learning. Expand
Improving Performance in Reinforcement Learning by Breaking Generalization in Neural Networks
TLDR
This paper finds that simply re-mapping the input observations to a high-dimensional space improves learning speed and parameter sensitivity and provides a simple approach to NN training that is easy to implement, and requires little additional computation. Expand
Concept Learning in Deep Reinforcement Learning
TLDR
It is shown that it is possible to learn concept-like representations by self-supervision, following an information-bottleneck approach, and that these representations accelerate the transference of skills by providing a prior that guides the policy optimization process. Expand
DDPG Agent to Swing Up and Balance Cart- Pole System
TLDR
This thesis studies state-of-the-art deep reinforcement learning algorithm, Deep Deterministic Policy Gradients, providing a theoretical comparison to other popular methods, an evaluation of its performance, identify its limitations and investigate future directions of research. Expand
Learning model-based strategies in simple environments with hierarchical q-networks
TLDR
The results show how a biologically inspired hierarchical learner can facilitate learning abstract rules to promote robust and flexible action policies in simplified training environments with clearly delineated optimal strategies. Expand
Deep Reinforcement Learning for Continuous Control
Reinforcement learning is a mathematical framework for agents to interact intelligently with their environment. In this field, real-world control problems are particularly challenging because of theExpand
Deep Reinforcement Learning for the Control of Robotic Manipulation: A Focussed Mini-Review
TLDR
This paper presents recent significant progress of deep reinforcement learning algorithms, which try to tackle the problems for the application in the domain of robotic manipulation control, such as sample efficiency and generalization. Expand
DEEP REINFORCEMENT LEARNING AGENTS
As deep reinforcement learning driven by visual perception becomes more widely used there is a growing need to better understand and probe the learned agents. Understanding the decision makingExpand
Developmental Reinforcement Learning through Sensorimotor Space Enlargement
TLDR
This work proposes a new type of transfer learning, inspired by the child development, where the sensorimotor space of an agent grows while it is learning a policy, and shows that searching first for a suboptimal solution in a subset of the parameter space, and then in the full space is helpful to bootstrap learning algorithms, and thus reach better performances in fewer episodes. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 36 REFERENCES
Reinforcement learning for robots using neural networks
TLDR
This dissertation concludes that it is possible to build artificial agents than can acquire complex control policies effectively by reinforcement learning and enable its applications to complex robot-learning problems. Expand
Reinforcement Learning: An Introduction
TLDR
This book provides a clear and simple account of the key ideas and algorithms of reinforcement learning, which ranges from the history of the field's intellectual foundations to the most recent developments and applications. Expand
Deep auto-encoder neural networks in reinforcement learning
TLDR
A framework for combining the training of deep auto-encoders (for learning compact feature spaces) with recently-proposed batch-mode RL algorithms ( for learning policies) is proposed and an emphasis is put on the data-efficiency and on studying the properties of the feature spaces automatically constructed by the deep Auto-encoder neural networks. Expand
Reinforcement learning can account for associative and perceptual learning on a visual decision task
TLDR
This work modeled performance on the basis of the readout of simulated responses of direction-selective sensory neurons in the middle temporal area of monkey cortex and found that changes can be accounted for using a reinforcement-learning rule to shape functional connectivity between the sensory and decision neurons. Expand
The Arcade Learning Environment: An Evaluation Platform for General Agents (Extended Abstract)
TLDR
The promise of ALE is illustrated by developing and benchmarking domain-independent agents designed using well-established AI techniques for both reinforcement learning and planning, and an evaluation methodology made possible by ALE is proposed. Expand
Learning Deep Architectures for AI
TLDR
The motivations and principles regarding learning algorithms for deep architectures, in particular those exploiting as building blocks unsupervised learning of single-layer modelssuch as Restricted Boltzmann Machines, used to construct deeper models such as Deep Belief Networks are discussed. Expand
Reinforcement learning for robot soccer
TLDR
Several variants of the general batch learning framework are discussed, particularly tailored to the use of multilayer perceptrons to approximate value functions over continuous state spaces, which are successfully used to learn crucial skills in soccer-playing robots participating in the RoboCup competitions. Expand
An object-oriented representation for efficient reinforcement learning
TLDR
Object-Oriented MDPs (OO-MDPs) are introduced, a representation based on objects and their interactions, which is a natural way of modeling environments and offers important generalization opportunities and a polynomial bound on its sample complexity is proved. Expand
A Neural Substrate of Prediction and Reward
TLDR
Findings in this work indicate that dopaminergic neurons in the primate whose fluctuating output apparently signals changes or errors in the predictions of future salient and rewarding events can be understood through quantitative theories of adaptive optimizing control. Expand
Neural Fitted Q Iteration - First Experiences with a Data Efficient Neural Reinforcement Learning Method
TLDR
NFQ, an algorithm for efficient and effective training of a Q-value function represented by a multi-layer perceptron, is introduced and it is shown empirically, that reasonably few interactions with the plant are needed to generate control policies of high quality. Expand
...
1
2
3
4
...