Talk:Outline of machine learning

Outlines High‑importance

	This article is an outline, a type of article that presents a list of articles or sub-topics related to its subject in a hierarchical form. For the standardized set of outlines on Wikipedia, see Wikipedia:Contents/Outlines. Outlines are within the scope of WikiProject Outlines, a collaborative effort to improve outlines on Wikipedia. For guidance on building and maintaining outlines, see Wikipedia:Outlines.OutlinesWikipedia:WikiProject OutlinesTemplate:WikiProject OutlinesOutlines articles
High	This article has been rated as High-importance on the importance scale.

Tip: Anchors are case-sensitive in most browsers.

This article links to one or more target anchors that no longer exist.

[[Recommender system#Hybrid recommender systems|Hybrid recommender systems]] The anchor (#Hybrid recommender systems) has been deleted by other users before.

Please help fix the broken anchors. You can remove this template after fixing the problems. | Reporting errors

Quick explanation of Wikipedia outlines

"Outline" is short for "hierarchical outline". There are two types of outlines: sentence outlines (like those you made in school to plan a paper), and topic outlines (like the topical synopses that professors hand out at the beginning of a college course). Outlines on Wikipedia are primarily topic outlines that serve 2 main purposes: they provide taxonomical classification of subjects showing what topics belong to a subject and how they are related to each other (via their placement in the tree structure), and as subject-based tables of contents linked to topics in the encyclopedia. The hierarchy is maintained through the use of heading levels and indented bullets. See Wikipedia:Outlines for a more in-depth explanation. The Transhumanist 18:24, 11 April 2017 (UTC)[reply]

Semi-protected edit request on 5 July 2017

This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.

I miss haveing the XGBoost algorithm mentioned underneath the decision tree section: https://wiki.riteme.site/wiki/Xgboost Maggyl (talk) 08:25, 5 July 2017 (UTC)[reply]

Not done: This is an outline about methods, not products or software libraries. The section "Machine learning methods" already includes links to both Gradient boosting machine (GBM) and Gradient boosted decision tree (GBRT), which XGBoost is but one exemplar of. Eggishorn (talk) (contrib) 18:20, 5 July 2017 (UTC)[reply]

Special:Permalink/807591251#Other

@The Transhumanist: I'm assuming that you copy/pasted all of the links in Category:Machine learning and its subcategories into the section linked in the title. A lot of those articles are about topics that are only tangentially related to machine learning; this is because a lot of editors tend to add certain categories to articles if the category topic is covered somewhere in an article despite the article topic not being relevant to the category.

In any event, you should comb through the entries in that section and remove the irrelevant ones. I made a first-pass attempt at this by removing the category and thumbnails that you added, along with fixing a typographical error, in these edits. Seppi333 (Insert 2¢) 23:16, 28 October 2017 (UTC)[reply]

@Seppi333: Thank you for the collaboration. The more the merrier. I'll be painstakingly placing each link into the body of the outline, tossing those that don't belong. Feel free to continue filtering. We can work from the "other" section, or those links can be placed here on the talk page and processed from here. Whatever you think is best. Either way is fine with me. By the way, you wouldn't by any chance know of a machine learning application we could use for this, would you? The Transhumanist 23:26, 28 October 2017 (UTC)[reply]

Haha, I wish. I'm fine with leaving those links in that section and moving them into the appropriate sections in the outline as time permits. I can probably work on filtering/sorting the links in that section more tomorrow; right now, I'm working on expanding the structural break article. Seppi333 (Insert 2¢) 23:43, 28 October 2017 (UTC)[reply]

@Seppi333: I've checked a bunch of 'em and those that seem to belong, I moved up into the parent see also section to wait for placement in the body of the outline. Those I weren't sure of I put in the section below on this talk page. The list compare tool in WP:AWB speeds things up as it can be applied to pull entire categories out of the list under "other" (to grab that list, you put it in a user sandbox page). The tool also has a filter to further speed up list processing. I used that filter to keep all the titles with "learning" in them, and loaded them into tabs for inspection. Turned out that they all are subtopics of machine learning, and so I moved them up into see also. I hope you find these tips useful. The Transhumanist 01:04, 29 October 2017 (UTC)[reply]

I unfortunately can't use AWB anymore due to some sort of glitch with the software; AWB used to work on my laptop up until about 1.5 years ago. I explained the issue here - Wikipedia talk:AutoWikiBrowser/Bugs#AutoWikiBrowser has stopped working - but no one seems to either want to, or possibly no one knows how to, address the issue I described there. Consequently, I'll have to manually check pages.

Seppi333 (Insert 2¢) 19:32, 29 October 2017 (UTC)[reply]

Not sure if these fall within the field of machine learning

(1+ε)-approximate nearest neighbor search - likely N
AAAI Conference on Artificial Intelligence - N
Absorbing Markov chain - N
Adaptive sampling - N
Affinity propagation - Y
AI@50 (had at least one lecture on machine learning) - N
Alexei A. Efros - N
Algorithmic composition - N
Angoss - N

— Preceding unsigned comment added by The Transhumanist (talk • contribs) 20:19, 28 October 2017 (UTC)[reply]

@The Transhumanist:I don't thing Angoss, Algorithmic composition, or "Alexei A. Efros" belong in the outline based upon those articles. There's only a tangential relationship to ML on the first two pages and the third one (Alexei) doesn't have a clear relationship to ML. "Absorbing Markov chain" is a probabilistic concept, not an ML one. Adaptive sampling is a mathematical/computational simulation model, as opposed to an ML algorithm/model, so it doesn't belong here either. (1+ε)-approximate nearest neighbor search is the only one I'm not 100% sure about since I don't have a strong background in computer science, but I'm going to say no to that one because the listed methods on that page for solving this optimization problem aren't classified as ML algorithms and don't mention ML.
~~AI@50 probably belongs because it's a conference about a major sub-field of machine learning (AI). Similar reasoning with AAAI Conference on Artificial Intelligence.~~ Affinity propagation is an ML algorithm that is probably most applicable to Outline of machine learning#Cluster analysis. Seppi333 (Insert 2¢) 19:27, 29 October 2017 (UTC)[reply]

Which is the parent field: AI or machine learning?

@Seppi333: We have an outline on both. So, which one is a branch of the other? The Transhumanist 20:06, 29 October 2017 (UTC)[reply]

My bad, seems my understanding of their relationship was backwards based upon an NVIDIA source (NVIDIA designs some of its GPUs for AI applications) and a Harvard Business Review article. AI is the broader topic. Seppi333 (Insert 2¢) 20:15, 29 October 2017 (UTC)[reply]

Split off artificial neural networks

@Seppi333: This is such a large subtopic, that I've split it off into its own outline at Wikipedia:WikiProject Outlines/Drafts/Outline of artificial neural networks. It will be moved to article space once it has been fleshed out. Feel free to help get it ready for prime time (article space). :) The Transhumanist 21:33, 29 October 2017 (UTC)[reply]

Sure thing. I'll work on it as time permits. I got sidetracked from a link I followed in this article (Uniform convergence in probability) which needed to be copyedited and also needs to be merged into Convergence of random variables and Proofs of convergence of random variables. It's a concept in asymptotic theory (statistics), so it provides a theoretical framework for certain types of statistical/machine learning models or hypothesis tests involving large sample sizes that require uniform convergence in probability as an assumption. If the uniform convergence article isn't merged into the 2 articles I tagged with merge templates, I don't think it belongs on this page unless asymptotic theory (statistics) and convergence of random variables are also included here. Seppi333 (Insert 2¢) 21:48, 29 October 2017 (UTC)[reply]

Encyclopedic?

Is it necessary to ask questions in an encyclopedia? -

"What type of thing is machine learning?"

Can't it just be stated?

--Mortense (talk) 18:34, 16 June 2020 (UTC)[reply]