Non-oblivious Strategy Improvement - Computer Science > Computer Science and Game TheoryReport as inadecuate




Non-oblivious Strategy Improvement - Computer Science > Computer Science and Game Theory - Download this document for free, or read online. Document in PDF available to download.

Abstract: We study strategy improvement algorithms for mean-payoff and parity games. Wedescribe a structural property of these games, and we show that thesestructures can affect the behaviour of strategy improvement. We show howawareness of these structures can be used to accelerate strategy improvementalgorithms. We call our algorithms non-oblivious because they rememberproperties of the game that they have discovered in previous iterations. Weshow that non-oblivious strategy improvement algorithms perform well onexamples that are known to be hard for oblivious strategy improvement. Hence,we argue that previous strategy improvement algorithms fail because they ignorethe structural properties of the game that they are solving.



Author: John Fearnley

Source: https://arxiv.org/







Related documents