Reinforcement Learning Embedded in Brains and Robots

Weber, Cornelius; Elshaw, Mark; Wermter, Stefan; Triesch, Jochen; Willmot, Christopher

doi:10.5772/5278

Cited by 9 publications

(6 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In selected states, the trainer advises the learner-agent changing the action to be performed in the environment. and how to get to states with higher values to reach the target by performing actions (Weber et al, 2008).…”

Section: Interactive Reinforcement Learningmentioning

confidence: 99%

Improving interactive reinforcement learning: What makes a good teacher?

et al. 2018

Self Cite

View full text Add to dashboard Cite

Interactive reinforcement learning has become an important apprenticeship approach to speed up convergence in classic reinforcement learning problems. In this regard, a variant of interactive reinforcement learning is policy shaping which uses a parent-like trainer to propose the next action to be performed and by doing so reduces the search space by advice. On some occasions, the trainer may be another artificial agent which in turn was trained using reinforcement learning methods to afterward becoming an advisor for other learner-agents.In this work, we analyze internal representations and characteristics of artificial agents to determine which agent may outperform others to become a better trainer-agent. Using a polymath agent, as compared to a specialist agent, an advisor leads to a larger reward and faster convergence of the reward signal and also to a more stable behavior in terms of the state visit frequency of the learner-agents. Moreover, we analyze system interaction parameters in order to determine how influential they are in the apprenticeship process, where the consistency of feedback is much more relevant when dealing with different learner obedience parameters.

show abstract

Section: Interactive Reinforcement Learningmentioning

confidence: 99%

Improving interactive reinforcement learning: What makes a good teacher?

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Doya proposed that unsupervised learning happens in the cortex and reinforcement learning in the basal ganglia [11]. Accordingly, the cortex pre-processes data to yield a representation that is suitable for reinforcement learning (RL) by the basal ganglia [12].…”

Section: Introductionmentioning

confidence: 99%

Reward-driven learning of sensorimotor laws and visual features

Kleesiek

Engel

Weber

et al. 2011

2011 IEEE International Conference on Development and Learning (ICDL)

Self Cite

View full text Add to dashboard Cite

Abstract-A frequently reoccurring task of humanoid robots is the autonomous navigation towards a goal position. Here we present a simulation of a purely vision-based docking behavior in a 3-D physical world. The robot learns sensorimotor laws and visual features simultaneously and exploits both for navigation towards its virtual target region. The control laws are trained using a two-layer network consisting of a feature (sensory) layer that feeds into an action (Q-value) layer. A reinforcement feedback signal (delta) modulates not only the action but at the same time the feature weights. Under this influence, the network learns interpretable visual features and assigns goaldirected actions successfully. This is a step towards investigating how reinforcement learning can be linked to visual perception.

show abstract

“…These values represent distances from [13,40] The learning algorithm is based on SARSA [13,14] and summarized as follows.…”

Section: Network Architecture and Learningmentioning

confidence: 99%

“…Reinforcement learning (RL) is a biologically supported learning paradigm [13,14,3], which allows an agent to learn through experience acquired by interaction with its environment. Reinforcement learning neural network architectures have an input layer, which represents the agent's current state, and an output layer, which represents the chosen action given a certain input.…”

Section: Introductionmentioning

confidence: 99%

Real-World Reinforcement Learning for Autonomous Humanoid Robot Charging in a Home Environment

Navarro

Weber

Wermter

2011

Towards Autonomous Robotic Systems

Self Cite

View full text Add to dashboard Cite

Abstract. In this paper we investigate and develop a real-world reinforcement learning approach to autonomously recharge a humanoid Nao robot [1]. Using a supervised reinforcement learning approach, combined with a Gaussian distributed states activation, we are able to teach the robot to navigate towards a docking station, and thus extend the duration of autonomy of the Nao by recharging. The control concept is based on visual information provided by naomarks and six basic actions. It was developed and tested using a real Nao robot within a home environment scenario. No simulation was involved. This approach promises to be a robust way of implementing real-world reinforcement learning, has only few model assumptions and offers faster learning than conventional Q-learning or SARSA.

show abstract

Reinforcement Learning Embedded in Brains and Robots

Cited by 9 publications

References 74 publications

Improving interactive reinforcement learning: What makes a good teacher?

Improving interactive reinforcement learning: What makes a good teacher?

Reward-driven learning of sensorimotor laws and visual features

Real-World Reinforcement Learning for Autonomous Humanoid Robot Charging in a Home Environment

Contact Info

Product

Resources

About