Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

13

FLAIRS
2003

favoriteEmaildiscussreport

141views Artificial Intelligence» more FLAIRS 2003»

Learning from Reinforcement and Advice Using Composite Reward Functions

13 years 6 months ago

Learning from Reinforcement and Advice Using Composite Reward Functions

Download ranger.uta.edu

1 Reinforcement learning has become a widely used methodology for creating intelligent agents in a wide range of applications. However, its performance deteriorates in tasks with sparse feedback or lengthy inter-reinforcement times. This paper presents an extension that makes use of an advisory entity to provide additional feedback to the agent. The agent incorporates both the rewards provided by the environment and the advice to attain faster learning speed, and policies that are tuned towards the preferences of the advisor while still achieving the underlying task objective. The advice is converted to “tuning” or user rewards that, together with the task rewards, define a composite reward function that more accurately defines the advisor’s perception of the task. At the same time, the formation of erroneous loops due to incorrect user rewards is avoided using formal bounds on the user reward component. This approach is illustrated using a robot navigation task.

Vinay N. Papudesi, Manfred Huber

Real-time Traffic

Artificial Intelligence | Faster Learning Speed | FLAIRS 2003 | Lengthy Inter-reinforcement Times | User Reward |

claim paper

Related Content

» Active Learning for Reward Estimation in Inverse Reinforcement Learning

» Principled Methods for Advising Reinforcement Learning Agents

» Evolution of reward functions for reinforcement learning

» Combining manual feedback with subsequent MDP reward signals for reinforcement learning

» ModelBased Average Reward Reinforcement Learning

» Using Temporal Neighborhoods to Adapt Function Approximators in Reinforcement Learning

» A kNN Based Perception Scheme for Reinforcement Learning

» Transfer of task representation in reinforcement learning using policybased protovalue fun...

» Learning to play Tetris applying reinforcement learning methods

Post Info
More Details (n/a)

Added	31 Oct 2010
Updated	31 Oct 2010
Type	Conference
Year	2003
Where	FLAIRS
Authors	Vinay N. Papudesi, Manfred Huber

Comments (0)