Berenji, H.R., Vengerov, D.: Advantages of cooperation between reinforcement learning agents in di#cult stochastic problems. In: Proc. of the Nineth IEEE International Conference on Fuzzy Systems (FUZZ-IEEE '00). (2000) Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Learning, Cooperation, and Coordination in Multi-Agent Systems - Berenji, Vengerov Self-citation (Vengerov) (Correct)
....initialized, it can still converge during learning to a locally but not globally optimal policy that ignores many important aspects of the environment. Cooperation among agents during learning is essential in directing the adjustment of policies in the globally most beneficial direction. In (Berenji and Vengerov, 2000a) attached in Appendix, we have presented analytical examples of how agents learning individually can get stuck with using severely suboptimal policies and how sharing experience between agents during learning can help to eliminate this problem. Our work on cooperative learning in multi agent ....
....whether it should be reinitialized and start learning from scratch. 3. By combining the so called actor critic reinforcement learning algorithm with fuzzy rule bases, the learning process of each agent is guaranteed to converge to optimality. A proof of this FRL convergence result is given in (Berenji and Vengerov, 2000b) 3.2 Learning algorithms In our previous work on single and multi agent learning, each agent was using the Qlearning algorithm for updating its behavioral policy. In this section we will first describe the Q learning algorithm with fuzzy state aggregation and then we will describe the new ....
[Article contains additional citation context not shown here]
Berenji, Hamid R. and Vengerov, David. (2000a) "Advantages of Cooperation Between Reinforcement Learning Agents in Difficult Stochastic Problems." In proceedings of the 9th IEEE International Conference on Fuzzy Systems (FUZZ-IEEE '00).
Adaptive Communication and Coordination in Multi-agent Systems.. - Vengerov (2000) Self-citation (Vengerov) (Correct)
....following model will be used to simulate the dynamics of continuous specialization in a robotic colony, which was described in the previous section. This model is an extension of the previously developed general framework for studying multi agent learning developed for NASA Ames Research Center (Berenji and Vengerov 1999, 2000). More details about the general model as well as simulation results and their analysis can be found in these papers. The space of possible areas of expertise is represented by a 2 D space, where each location represents a certain degree of involvement in performing each task. For simplicity, this ....
....by averaging individual estimates, agents can evaluate more accurately the values of different types of opportunities. In this case, the global information can consist of more precise estimates of expected lifetime and reward for observed opportunities. This scenario was simulated and analyzed in (Berenji and Vengerov 1999, 2000). A more complicated and a more interesting scenario arises when the values of opportunities considered by an agent are not intrinsic but also depend on the actions of other agents. For example, a large number of agents clustered around a certain opportunity might indicate that the opportunity is ....
[Article contains additional citation context not shown here]
Berenji, Hamid R. and Vengerov, David (2000) "Advantages of Cooperation Between Reinforcement Learning Agents in Difficult Stochastic Problems." Accepted for publication in the 9th IEEE International Conference on Fuzzy Systems (FUZZ-IEEE '00). Available electronically from http://www.stanford.edu/~vengerov.
Exchanging Advice and Learning to Trust - Lus Nunes And (2003) (Correct)
No context found.
Berenji, H.R., Vengerov, D.: Advantages of cooperation between reinforcement learning agents in di#cult stochastic problems. In: Proc. of the Nineth IEEE International Conference on Fuzzy Systems (FUZZ-IEEE '00). (2000)
Cooperative Learning Using Advice Exchange - Lus Nunes Eugnio (2003) (Correct)
No context found.
H. R. Berenji, D. Vengerov. Advantages of Cooperation Between Reinforcement Learning Agents in Difficult Stochastic Problems. 9th IEEE International Conference on Fuzzy Systems (FUZZ-IEEE '00), 2000
V-Lab® - A Virtual Laboratory for Autonomous.. - El-Osery, Burge.. (2002) (Correct)
No context found.
Berenji, H.R. and Vengerov, D. "Advantages of cooperation between reinforcement learning agents in difficult stochastic problems," Proc. of the 9 th IEEE International Conference on Fuzzy Systems, Vol.2, pp. 871--876, 2000.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC