Talk:Neural network (machine learning)

icon
Archives (Index)

This page is archived by ClueBot III.

"Failed Verification" under Computational Power

Why does this sentence have a failed verification tag?

Further, the use of irrational values for weights results in a machine with super-Turing power.[failed verification]

The second citation appears to clearly support the claim made:

If the weights are rational numbers, then networks are equivalent to Turing machines [25], [26]; that is, they are no more powerful than standard digital computers. In the general case, weights are real numbers with infinite precision. Then the networks have been shown to be strictly more powerful than Turing machines [27].

Perhaps there is some confusion regarding rational approximations of real numbers with finite precision and "real" real numbers with infinite precision? GPeckman (talk) 00:14, 20 March 2026 (UTC)[reply]

Possible one-source reliance in History section

I think the history section may be largely derived from Jürgen Schmidhuber's "Annotated History of Modern AI and Deep Learning"; a vast majority of the sources in the section are either primary sources (links to research papers) or links to Schmidhuber's work. (I think I muddied the waters here a bit by copying over content from other sections which use separate sources, admittedly; Here's a link to the version before my rewriting attempt for reference.) I strongly suspect, though I have not proven it, that the "references" to primary sources are only included because Schmidhuber uses them.

There are some other suspicious things, like this typo in citation which matches exacty with a typo in the references for Schmidhuber's work; his article mistakenly cited Amari's 1967 work as "A theory of adaptive pattern classifier" rather than "A theory of adaptive pattern classifiers", which is copied over to the old version of the Wikipedia article.

I would need to do more research to figure out the exact similarities in content, but I'm leaving this here for future reference, and to remind myself (and other people) to update the history section to include a broader variety of tertiary sources. ARoseThorn (talk) 15:28, 27 March 2026 (UTC)[reply]

Well, it seems I've stumbled into a discussion that already happened in 2024; I didn't want to go down this rabbit hole, but regardless I think the section really does need a lot of clean up for readability. ARoseThorn (talk) 20:55, 27 March 2026 (UTC)[reply]

66:6F:B2:59:DB:76 ~2026-21252-13 (talk) 18:38, 6 April 2026 (UTC)[reply]

Suggestion to Expand Coverage of Recent Advances

The article provides a comprehensive overview of neural networks, including their history, architecture, and applications. However, I would like to suggest expanding the section on recent developments. Including more detailed information on transformers, generative AI, and large language models could enhance the article’s relevance and reflect current advancements in machine learning. Additionally, adding simplified diagrams or beginner-friendly explanations may improve accessibility for new learners.

Thank you to all contributors for maintaining this valuable resource.

Dikshitha Reddy Vanga (talk) 05:38, 12 April 2026 (UTC)[reply]

AI-generated (May 2026)

I removed {{AI-generated}} and replaced it with targeted tags on the user's contributions to the article, which require human review. The original note on the maintenance template was this March 2026 edit; note WP:AISIGNS - outline-like structure about challenges, superficial analyses, vocab distro typical of LLM output, etc. Wracking talk! 19:48, 17 May 2026 (UTC)[reply]

Content Disclaimer

Informasi ini disarikan dari Wikipedia dan disajikan kembali untuk tujuan edukasi. Konten tersedia di bawah lisensi CC BY-SA 3.0. Kami tidak bertanggung jawab atas ketidakakuratan data yang bersumber dari kontribusi publik tersebut.

  1. The information displayed on this website is sourced in part or in whole from Wikipedia and has been adapted for the purpose of restating it. We strive to provide accurate and relevant information, however:
  2. There is no guarantee of absolute accuracy. Wikipedia is an open, collaborative project that can be edited by anyone, so information is subject to change.
  3. It is not intended to constitute professional advice. The content displayed is for informational and educational purposes only. For important decisions (e.g., medical, legal, or financial), please consult a professional.
  4. Content copyright. Wikipedia is licensed under the Creative Commons Attribution-ShareAlike License (CC BY-SA). This means that content may be reused with appropriate attribution and shared under a similar license.
  5. Responsible use. Any risk arising from the use of information from this website is entirely the responsibility of the user.