Talk:Transfer learning

Untitled

The citations are missing references.

possible COI issue

This article seems to promote the work of Danny Silver and Rich Caruana. The author who created this article and wrote most of the text is named Dsilver, which makes me a bit suspicious that Danny Silver wrote this article to promote his own work. Lisatwo 05:59, 21 July 2007 (UTC)

As additional evidence, there is a professor in the real world named Daniel L. Silver who works on machine learning. DSilver has only contributed to articles concerning machine learning (including this article) and has signed one of his contributions as "DLSilver" (note the middle initial). Lisatwo 06:04, 21 July 2007 (UTC)

I concur

and have subsequently nominated the page for speedy deletion in sight of it's spam-like general tone.--Tsboncompte 21:17, 4 August 2007 (UTC)

It seems like some anonymous IP deleted a reference, thinking that this was the cause of the COI issue. However, the token reference was a conference paper that may have relevant information. I'm going to list it here in case anyone wants to go read and understand it well enough to add to the main article. The reference is:

Reid, M. D. (2004-07-30). "Improving Rule Evaluation Using Multitask Learning". Proceedings of the 14th International Conference on ILP. Retrieved 2006-12-26. {{cite conference}}: Check date values in: |date= (help); Unknown parameter |booktitle= ignored (|book-title= suggested) (help)

I think it would be unwise to throw out the entire article, since it seems to talk about a legitimate scientific idea. Lisatwo 01:31, 5 August 2007 (UTC)

Yes, it seems like that, doesn't it? But this smells more like marketing an idea just for the sake of marketing, to me. All this guy is doing is putting new words together and giving them specialized definitions, when simpler words already exist that describe the same phenomena. "Transfer learning"? Why doesn't he just call it intuition? Xaxafrad 05:29, 6 August 2007 (UTC)

The text of this article was an almost exact copy of [http://iitrl.acadiau.ca/itws05/overview.htm] apparently written by Danny Silver (web page says "Last updated 12/01/2005 - Danny Silver"). The website was mentioned in the article. I infer that Dr. Silver contributed this text here, unaware of Wikipedia's COI and other contribution guidelines. But whatever the case, the text needed to be removed.

I believe it is a legitimate topic for a wikipedia article, and I've rewritten a stub on the topic from scratch, with sources carefully cited. I've removed the COI warning, request for wikifying, and request for source citation. I don't know how to remove the request for deletion, as I'm not familiar with that process. -Agyle 07:34, 6 August 2007 (UTC)

Inductive Transfer or Transfer Learning

One question I have is whether "transfer learning" is a more common term for "inductive transfer." I think it is, but I'm not an expert, so I don't know for sure, and I don't know if there are subtle differences in the meanings. As such, I'm leaving it under this title, and will redirect "transfer learning" to "inductive transfer." -Agyle 07:36, 6 August 2007 (UTC)

@Agyle: I've never even heard of transfer learning being called inductive transfer. Indeed, when I search on Google Scholar for "transfer learning" + "machine learning" versus "inductive transfer" + "machine learning" I get 10x as many hits. I vote to rename the article Transfer Learning and redirect Inductive Transfer there. Richard☺Decal (talk) 15:16, 18 April 2017 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Transfer learning. 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/20070801120743/http://cpms.byu.edu/springresearch/abstract-entry?id=861 to http://cpms.byu.edu/springresearch/abstract-entry?id=861

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

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) 04:32, 18 September 2017 (UTC)

Deep Learning

Deep Learning is not mentioned yet. Maybe we should also write something about transfer learning with neural networks since transfer learning has become quite important in this area. From the top of my head I could name a long list of papers about that. Furthermore there are many checkpoints publicly available. (caffe model zoo, google mobileNet, etc.) — Preceding unsigned comment added by Valmendil (talk • contribs) 21:43, 21 September 2017 (UTC)

http://ruder.io/transfer-learning/index.html#learningdomaininvariantrepresentations presents some of the papers about the topic. it's worth reading I think. And most importantly it contains all the important links to the relevant papers https://medium.com/nanonets/nanonets-how-to-use-deep-learning-when-you-have-limited-data-f68c0b512cab presents some ideas as well

I propose to write something similar to how they present transfer learning

I agree that deep learning should be mentioned in this article, as it has become an important current usage of the concept of transfer learning. Showeropera (talk) 22:26, 30 January 2018 (UTC)

Gg

Ĝ 185.203.53.249 (talk) 15:10, 15 December 2022 (UTC)