4446 search results - page 300 / 890 » Learning Observer Agents |
189
click to vote
ICML
2003
IEEE
16 years 7 months ago
2003
IEEE
Learning in many multi-agent settings is inherently repeated play. This calls into question the naive application of single play Nash equilibria in multi-agent learning and sugges...
174
click to vote
ATAL
2003
Springer
15 years 12 months ago
2003
Springer
Much emphasis in multiagent reinforcement learning (MARL) research is placed on ensuring that MARL algorithms (eventually) converge to desirable equilibria. As in standard reinfor...
201
click to vote
CEEMAS
2003
Springer
15 years 12 months ago
2003
Springer
Modeling learning agents in the context of Multi-agent Systems requires an adequate understanding of their dynamic behaviour. Usually, these agents are modeled similar to the di�...
173
click to vote
AAAI
2007
15 years 9 months ago
2007
Transfer learning is the ability of an agent to apply knowledge learned in previous tasks to new problems or domains. We approach this problem by focusing on model formulation, i....
179
click to vote
ATAL
2010
Springer
15 years 6 months ago
2010
Springer
PAC-MDP algorithms approach the exploration-exploitation problem of reinforcement learning agents in an effective way which guarantees that with high probability, the algorithm pe...