Revision as of 11:28, 8 August 2018 edit 129.125.178.74 (talk) →Goals: There is an explicit advantage of anytime algorithms that was not mentioned: for a fixed computing duration, the actual performance in the application ___domain becomes better. ← Previous edit		Revision as of 03:59, 9 August 2018 edit undo GrapefruitSculpin (talk \| contribs) Extended confirmed users, Pending changes reviewers, Rollbackers 4,782 edits Reverted 1 edit by 129.125.178.74 (talk): Addition of unsourced content. Please see: WP:WHYCITE + WP:PROVEIT. . (TW) Tag: Undo Next edit →
Line 13: Anytime algorithms are designed so that it can be told to stop at any time and would return the best result it has found so far.<ref name="Grass"/> This is why it is called an interruptible algorithm. Anytime algorithms also maintain the last result, so that if they are given more time, they can continue from where they left off to obtain an even better result.<ref name="Grass"/> ~~An interesting consequence of the anytime paradigm is that the algorithm~~ will automatically yield a better response for a fixed computing time frame as processing power increases over the years due to technological advances. In the same amount of real-world time, e.g., one second, the search depth can be deeper, increasing the likelihood of returning better results. Fixed-condition terminating algorithms, on the other hand, would give exactly the same response as before, but faster, not better. == Decision trees ==

Anytime algorithm: Difference between revisions