Granular value-function approximation for road network traffic control


	BibTex \| RIS \| RefWorks Download

, , ,

The research discussed in this paper aims at developing fast stable learning agents for large-scale complex systems including network traffic signal control systems. The control system is based on reinforcement learning (RL), an important research area in distributed AI with a wide area of applications including real-time control. RL-based control may also be suitable for distributed domains that are subject to time and environmental contingencies. Based on this assumption, the goal in this paper is to investigate ways to make RL excel at on-line, continuous state and action space tasks by incorporating the concept of fuzzy granulation as (powerful) function approximation tool: we argue why this may strongly improve the learning speed of the algorithm. The potential implications of this research are better running times, allowing us to consider much larger problem sizes.

Conference: International Conference on Networking, Sensing and Control - ICNSC , 2010

DOI: 10.1109/ICNSC.2010.5461556

Cumulative Annual

View Publication

( ieeexplore.ieee.org )

Keywords (11)

References (25)

Reinforcement Learning: A Survey (Citations: 2036)

Delayed Reinforcement, Fuzzy Q-Learning and Fuzzy Logic Controllers (Citations: 25)

Between MDPs and Semi-MDPs: A Framework for Temporal Abstraction in Reinforcement Learning (Citations: 495)

Reinforcement learning in neurofuzzy traffic signal control (Citations: 28)

Signal Control for Oversaturated Intersections Using Fuzzy Logic (Citations: 9)


	The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.