Talk:Beam search

Norvig beam search

Paradigms of Artificial Intelligence programming has a clear explanation of Beam Search (with Lisp code):

http://books.google.com/books?id=X4mhySvjqUAC&pg=PA196&lpg=PA196&dq=beam+search+in+paradigms+of+artificial+intelligence+programming&source=web&ots=20BH2lB3sc&sig=udBIO2Qeg8awaySgwCu7gDEObY4&hl=en&sa=X&oi=book_result&resnum=1&ct=result#PPA196,M1 Jwinter (talk) 18:26, 26 June 2008 (UTC)jwinter

source code for beam search in same book (from author's site):

http://norvig.com/paip/search.lisp 217.132.197.231 (talk) 22:27, 6 July 2008 (UTC)

Not happy about discussion of variable beam width

Paragraph currently reads as follows: The beam width can either be fixed or variable. In a fixed beam width, a maximum number of successor states is kept. In a variable beam width, a threshold is set around the current best state. All states that fall outside this threshold are discarded. Thus, in places where the best path is obvious, a minimal number of states is searched. In places where the best path is ambiguous, many paths will be searched. The wording is problematic. A threshold is not set around a state; a threshold is a left wall or right wall rather than a pair of walls. Also, the paragraph seems to be confusing fixed width with maximum width. Finally, the paragraph doesn't make it clear how the width is chosen. I suspect that there are a variety of ways. One way, described in the book reference above, is what the author calls "iterative widening" (though he emphasizes that this is not a standard term): the beam width is started at a minimum level, and if the search fails, the width is increased and the search repeated. I'm sure there are other ways of dynamically setting the beam width, however. I think it would be more accurate to simply remove everything but the first sentence of the paragraph, which is probably what I'll do. -AlanUS (talk) 20:12, 1 March 2011 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified 2 external links on Beam search. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20080516202406/http://www.ijcai.org/papers/0596.pdf to http://www.ijcai.org/papers/0596.pdf
• Added archive https://web.archive.org/web/20060618020933/http://acl.ldc.upenn.edu:80/J/J03/J03-1005.pdf to http://acl.ldc.upenn.edu/J/J03/J03-1005.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 08:55, 29 October 2016 (UTC)

Proposed addition on node-wise beam search

I am a co-author of the cited paper, so I am requesting review by an uninvolved editor.

Please add the following paragraph in the “Variants” section, immediately after the paragraph about stochastic beam search and before “Other variants are flexible beam search and recovery beam search”:

Another approach to mitigating local optima is node-wise beam search, which retains up to a fixed number of candidate paths for each graph node rather than retaining the best candidates globally at each search depth. This allows a wider range of regions in the search space to remain under consideration and has been applied to active perception and informative path planning in mobile robotics.

The paper describes standard beam search as retaining the top candidates globally at each depth and introduces node-wise beam search, which instead retains candidates per graph node to mitigate local optima and improve exploration.

Kaixqu (talk) 16:14, 12 July 2026 (UTC)

Kaixqu (talk) 16:14, 12 July 2026 (UTC)

References

Reply 26-JUL-2026

  Edit request declined  

• A page number or quote from the source was not provided with the request.
• Additionally, please provide a WP:SECONDARY source.

Regards,  Spintendo  20:50, 26 July 2026 (UTC)