Agent-Based System with Learning Capabilities for Transport Problems

被引：0

作者：

Sniezynski, Bartlorniej ^{[1
]}

Kozlak, Jaroslaw ^{[1
]}

机构：

[1] AGH Univ Sci & Technol, Fac Elect Engn Automat Comp Sci & Elect, Dept Comp Sci, PL-30059 Krakow, Poland

来源：

COMPUTATIONAL COLLECTIVE INTELLIGENCE: TECHNOLOGIES AND APPLICATIONS, PT II: THIRD INTERNATIONAL CONFERENCE, ICCCI 2011 | 2011年 / 6923卷

关键词：

agent-based system; machine learning; transport problem;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this paper we propose an agent architecture with learning capabilities and its application to a transportation problem. The agent consists of the several modules (control, execution, communication, task evaluation, planning and social) and knowledge bases to store information and learned knowledge. The proposed solution is tested on the PDPTW. Agents using supervised and reinforcement learning algorithms generate knowledge to evaluate arriving requests. Experimental results show that learning increases agent performance.

引用

页码：100 / 109

页数：10

共 19 条

[1] Airiau S., 2008, P ALAMAS ALAG WORKSH
[2] [Anonymous], 2014, C4. 5: programs for machine learning
[3] The simulated trading heuristic for solving vehicle routing problems
Bachem, A
Hochstattler, W
Malich, M
[J]. DISCRETE APPLIED MATHEMATICS, 1996, 65 (1-3) : 47 - 72
[4] Cohen WW, 1995, MACHINE LEARNING
[5] Cornite F.D., 2005, JAVA PLATFORM REINFO
[6] Cooperative transportation scheduling: An application domain for DAI
Fischer, K
Muller, JP
Pischel, M
[J]. APPLIED ARTIFICIAL INTELLIGENCE, 1996, 10 (01) : 1 - 33
[7] Frank E., 1999, Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations
[8] Gehrke JD, 2008, LECT NOTES COMPUT SC, V5103, P692, DOI 10.1007/978-3-540-69389-5_77
[9] Kozlak J, 2007, LECT NOTES COMPUT SC, V4488, P872
[10] Kozlak J, 2006, FUND INFORM, V71, P27

← 1 2 →