http://robowiki.net/w/index.php?title=Thread:Talk:DrussGT/Understanding_DrussGT/Reason_behind_using_Manhattan_distance/reply_(4)&feed=atom&action=historyThread:Talk:DrussGT/Understanding DrussGT/Reason behind using Manhattan distance/reply (4) - Revision history2024-03-28T22:53:49ZRevision history for this page on the wikiMediaWiki 1.34.1http://robowiki.net/w/index.php?title=Thread:Talk:DrussGT/Understanding_DrussGT/Reason_behind_using_Manhattan_distance/reply_(4)&diff=54763&oldid=prevMN: Reply to Reason behind using Manhattan distance2018-08-28T00:28:37Z<p>Reply to <a href="/wiki/Thread:Talk:DrussGT/Understanding_DrussGT/Reason_behind_using_Manhattan_distance" title="Thread:Talk:DrussGT/Understanding DrussGT/Reason behind using Manhattan distance">Reason behind using Manhattan distance</a></p>
<p><b>New page</b></p><div>I have 2 hypotheses:<br />
<br />
- Manhattan distance is more tolerant to noise than Euclidean distance. Squaring a dimension amplifies noise.<br />
<br />
- [https://en.wikipedia.org/wiki/Curse_of_dimensionality Curse of dimensionality]. Euclidean distance behaves oddly at high dimensions.</div>MN