Changes between Version 100 and Version 101 of OpponentModels

Timestamp:

05/10/11 14:59:35 (14 years ago)

Author:

mark

Comment:

--

OpponentModels

@@ -11 +11 @@
 4.c. Learning Utility curves[[BR]]
 4.d. Learning Value Ordering[[BR]]
-4.e. Learning Weight Ordering[[BR]]
-4.f. Learning Time Left[[BR]]
-5. Future Work[[BR]]
-6. Conclusion[[BR]]
+4.e. Learning Issue Ordering[[BR]]
+5. Evaluating Opponent Models[[BR]]
+5. Opponent Model Based Strategies: Exploiting opponent models[[BR]]
+6. Future Work[[BR]]
+7. Conclusion[[BR]]
 
 == Papers ==
@@ -464 +465 @@
 ||'''Author(s)'''||K. Hindriks, C.M. Jonker, D. Tykhonov||
 ||'''Cited'''||-||
-||'''Subject(s)'''||||
-||'''Summary'''||||
-||'''Relevance'''||||
+||'''Subject(s)'''||Nice Mirroring Strategy using Bayesian Learning||
+||'''Summary'''||Opponent models can aid in preventing exploitation, by determining the type of move of the opponent (selfish, [[BR] unfortunate, concession, nice, fortunate, silent), and by taking the opponents preferences into account to[[BR] increase the chance of acceptation. The mirror strategy mirrors the behaviours of the opponent, based on a [[BR]classification of the opponent move. Nice MS does the same, but adds a nice move, which is a move which only [[BR]increases the opponents utility without decreasing ours. Overall the strategy is shown to be effective by [[BR]comparing the result of first testing the strategy against a random agent, and then the other agents. Also, the[[BR] distance to a Kalai-Smorodinsky solution and the distance to the Nash Point is used as a metric. For [[BR]future work the exploitability of MS should be researched.||
+||'''Relevance'''||8, interesting application of opponent modelling ||
 ||'''Bibtex'''||[http://scholar.google.nl/scholar.bib?q=info:BtssqMir4RcJ:scholar.google.com/&output=citation&hl=nl&as_sdt=0,5&ct=citation&cd=2 Link]||
 ||'''Cites seen'''||Yes||