Filters








5,623 Hits in 6.6 sec

INCREMENTAL LEARNING OF PROCEDURAL PLANNING KNOWLEDGE IN CHALLENGING ENVIRONMENTS

Douglas J. Pearson, John E. Laird
2005 Computational intelligence  
One class of existing learners, reinforcement learners, typically employ weak learning methods to directly modify an agent's execution knowledge.  ...  This research investigates the hypothesis that by limiting an agent to procedural access to symbolic planning knowledge, the agent can combine the powerful, knowledge-intensive learning performance of  ...  Failures during execution can be detected by comparing the agent's planned behavior to the agent's actual behavior in the environment during execution.  ... 
doi:10.1111/j.1467-8640.2005.00280.x fatcat:igl2sbkzynfstpw2hjkea7clie

Negotiation Planning Of Autonomous Agents In Multi-agent Environment

N. Sadeghpoor, A. Rahati
2014 Journal of Mathematics and Computer Science  
Negotiation is usually carried out during the plan execution, however, it can be considered during the planning stage, as in real life.  ...  In other words, an exact plan of arguments and their order should be prepared.  ...  In section 3, we decribe our nondeterministic negotiation planning algorithm that can overcome this shortcome by modeling incomplete knowedge through some sensing action and specific way of knowledge modeling  ... 
doi:10.22436/jmcs.010.04.05 fatcat:irsnhqf3mvfj3n2xtav3audaty

S-assess

Scott A. Wallace
2005 Proceedings of the fourth international joint conference on Autonomous agents and multiagent systems - AAMAS '05  
In this paper we present a framework that brings validation techniques out of the laboratory and uses them to monitor and constrain an agent's behavior concurrent with task execution.  ...  Developing and testing intelligent agents is a complex task that is both time-consuming and costly.  ...  During validation, domain experts and knowledge engineers examine the agent's behavior in each test scenario looking for errors and correcting the agent's knowledge as required.  ... 
doi:10.1145/1082473.1082512 dblp:conf/atal/Wallace05 fatcat:i7ii667eavg3zh7iozufgknvu4

Avoiding Negative Side Effects due to Incomplete Knowledge of AI Systems [article]

Sandhya Saisubramanian and Shlomo Zilberstein and Ece Kamar
2021 arXiv   pre-print
Due to the limited fidelity of its model, an agent's actions may have unexpected, undesirable consequences during execution.  ...  Learning to recognize and avoid such negative side effects of an agent's actions is critical to improve the safety and reliability of autonomous systems.  ...  Due to the practical limitations of data collection and model specification, agents operating in the open world often rely on incomplete knowledge of their target environment which may lead to unexpected  ... 
arXiv:2008.12146v3 fatcat:np263lgag5ejpaywdq5o2o2lqe

Avoiding Negative Side Effects Due to Incomplete Knowledge of AI Systems

Sandhya Saisubramanian, Shlomo Zilberstein, Ece Kamar
2022 The AI Magazine  
Due to the limited fidelity of its model, an agent's actions may have unexpected, undesirable consequences during execution.  ...  Learning to recognize and avoid such negative side effects (NSEs) of an agent's actions is critical to improve the safety and reliability of autonomous systems.  ...  Due to the practical limitations in data collection and model specification, agents operating in the open world often rely on incomplete knowledge of their target environment which may lead to unexpected  ... 
doi:10.1609/aimag.v42i4.7390 fatcat:z4bybhdzofcgxcagvmxxv2vjpy

A Methodology for Developing Self-explaining Agents for Virtual Training [chapter]

Maaike Harbers, Karel van den Bosch, John-Jules Meyer
2010 Lecture Notes in Computer Science  
By using an agent programming language in which declarative aspects of an agent's reasoning process are explicitly represented, explanations revealing the underlying motivations for agents' actions can  ...  In this paper, a methodology for developing self-explaining agents in virtual training systems is proposed, resulting in agents that can explain their actions in terms of beliefs and goals.  ...  ACKNOWLEDGMENTS This research has been supported by the GATE project, funded by the Netherlands Organization for Scientific Research (NWO) and the Netherlands ICT Research and Innovation Authority (ICT  ... 
doi:10.1007/978-3-642-13338-1_10 fatcat:mtkzxs6q4newla3yl32evuccvy

Agent-Based Evolutionary Model for Knowledge Acquisition in Dynamical Environments [chapter]

Wojciech Froelich, Marek Kisiel-Dorohinicki, Edward Nawarecki
2006 Lecture Notes in Computer Science  
In the discussed variant of the model, each agent uses reinforcement learning, and the obtained knowledge is represented as the set of simple decision rules.  ...  The basic idea of the approach proposed in this paper is to apply multiagent paradigm in order to enable the integration and co-operation of different knowledge acquisition and representation techniques  ...  Further, the tests had been executed, during which, the agent's learning based on the suggested model had been performed.  ... 
doi:10.1007/11758532_109 fatcat:wlyptzpptzfhlbv6wammjffry4

Programming and simulation of quantum search agents

Matthias Klusch, Rene Schubotz
2007 2007 IEEE International Conference on Systems, Man and Cybernetics  
The extension of classical agents by the ability to perform quantum computation and communication provides an efficient and secure solution to applications such as information search and service matchmaking  ...  Finally, we present preliminary results of the comparative evaluation of its implementation using different quantum programming languages and simulators.  ...  The KB consists of the agent's world model, the agent's local goals, the behavioural knowledge, the local planning knowledge and the cooperative planning knowledge.  ... 
doi:10.1109/icsmc.2007.4413701 dblp:conf/smc/KluschS07 fatcat:kvvgenurizb6fee6old5pvbvpm

Simulating a Human Cooperative Problem Solving [chapter]

Alexandre Pauchet, Amal El Fallah Seghrouchni, Nathalie Chaignaud
2007 Lecture Notes in Computer Science  
We are interested in understanding and simulating how humans elaborate plans in situations where knowledge is incomplete and how they interact to obtain missing information.  ...  The system BDIggy we propose, is a concurrent implementation of the planning model and the interaction model through the BDI concept.  ...  The execution module performs the actions of an intention. During parallel planning, the plan interpreter manages as many desire and intention stacks as there are parallel plans.  ... 
doi:10.1007/978-3-540-75254-7_23 fatcat:6mccc2ncbncq7huydoj2y7x7ta

Programming and simulation of quantum search agents

Matthias Klusch, René Schubotz
2007 Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems - AAMAS '07  
The extension of classical agents by the ability to perform quantum computation and communication provides an efficient and secure solution to applications such as information search and service matchmaking  ...  Finally, we present preliminary results of the comparative evaluation of its implementation using different quantum programming languages and simulators.  ...  The KB consists of the agent's world model, the agent's local goals, the behavioural knowledge, the local planning knowledge and the cooperative planning knowledge.  ... 
doi:10.1145/1329125.1329230 dblp:conf/atal/KluschS07 fatcat:dmkpt4td5jajpkjrfifudot53a

Modeling reactive behaviour in vertically layered agent architectures [chapter]

Jörg P. Müller, Markus Pischel, Michael Thiel
1995 Lecture Notes in Computer Science  
The paper focusses on the lower levels of the architecture which provide reactivity, incorporate procedural knowledge, and which connect the cooperation and planning layers with the outside world.  ...  We claim that the lower system layers are likely to become a control bottleneck in vertically layered architectures, and that very careful modeling is required to produce the desired agent behaviour.  ...  The work presented in this paper has been supported by the German Ministry of Research and Technology under grant ITW9104.  ... 
doi:10.1007/3-540-58855-8_17 fatcat:aogyaquee5exlgsddklkan5hga

GOAL-DRIVEN AUTONOMY FOR RESPONDING TO UNEXPECTED EVENTS IN STRATEGY SIMULATIONS

Matthew Klenk, Matt Molineaux, David W. Aha
2012 Computational intelligence  
To operate autonomously in complex environments, an agent must monitor its environment and determine how to respond to new situations.  ...  By employing goal reasoning, ARTUE outperforms an off-line planner and a discrepancy-based replanner on scenarios requiring reasoning about unobserved objects and facts and on scenarios presenting opportunities  ...  Matthew Klenk performed this research while supported by an NRC postdoctoral fellowship.  ... 
doi:10.1111/j.1467-8640.2012.00445.x fatcat:xdvsb5bfl5clzhmll25mqy3uiy

Study about Law Multi-Issue Automatic Negotiation Method Based on Artificial Intelligence and Multiagent Evolutionary Algorithm

Yu-Ting Hsu, Cheng-Yong Liu, Sang-Bing Tsai
2021 Mobile Information Systems  
Each agent in the MAS works independently of the other, and they have all the characteristics of an agent system.  ...  According to this problem, the autonegotiation solution process and corresponding model are designed. In addition, a new type of solution is proposed for multiple legal issues.  ...  But agent's knowledge shows that cooperating with other agents can improve workload and produce better results.  ... 
doi:10.1155/2021/7236900 fatcat:v2eys6ysrnaepbfgp2gz7xlhri

Architectures by Design: The Iterative Development of an Integrated Intelligent Agent [chapter]

Nick Hawes
2009 Research and Development in Intelligent Systems XXVI  
We present an instance of this design methodology applied to the development of an architecture that integrates anytime deliberative capabilities with reactive behaviours and goal management.  ...  Iterations of the design are implemented and evaluated in the computer game Unreal Tournament.  ...  Execution time: The amount of time an agent spends executing plans during a single game. • Wasted execution time: The amount of time an agent spends executing plans to achieve a goal after the goal is  ... 
doi:10.1007/978-1-84882-983-1_28 dblp:conf/sgai/Hawes09 fatcat:dulmxhgtpfaardkbplr5zbyeg4

Interactive Learning of Grounded Verb Semantics towards Human-Robot Communication

Lanbo She, Joyce Chai
2017 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)  
The proposed approach uses reinforcement learning to allow the robot to acquire an optimal policy for its question-asking behaviors by maximizing the long-term reward.  ...  The rich interaction between teachers and students that is considered important in learning new skills has not yet been explored.  ...  Misra and colleagues for providing the evaluation data, and the anonymous reviewers for valuable comments.  ... 
doi:10.18653/v1/p17-1150 dblp:conf/acl/SheC17 fatcat:2nk6x3f3bbbftgd6x6ieqt7jle
« Previous Showing results 1 — 15 out of 5,623 results