Talk:Deep learning

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Energy consumption[edit]

I am considering adding information regarding power consumption. Currently I see two directions: low-power inference for embedded devices[1] and low-power training for huge networks[2]. Please let me know if you know of more credible and/or overview sources discussing the concerns. p.s. This may be a discussion for another page, e.g. ML or ANNs. Cheater no1 (talk) 13:16, 13 November 2020 (UTC)[reply]

Planned slow motion cleanup of reference spamming and overciting[edit]

I plan to attempt a slow motion cleanup of reference spamming. The typical case is an obscure non-notable paper where it does a poor job of supporting the items which cited it, particularly where the material is still cited without it. When all of the references looked equally good/applicable on overcited items, I left them all in. Sincerely, North8000 (talk) 16:29, 7 March 2022 (UTC)[reply]

  • Completed through the first image and lead. North8000 (talk) 20:20, 7 March 2022 (UTC)[reply]
  • Completed through the first 5 paragraphs of the overview section. North8000 (talk) 22:43, 7 March 2022 (UTC)[reply]
  • Completed through end of overview section. North8000 (talk) 02:54, 8 March 2022 (UTC)[reply]
  • Complete through interpretation section. That section still has some over-cited items, but the remaining ones all look useful. North8000 (talk) 17:56, 14 March 2022 (UTC)[reply]
  • Complete through 1st 13 paragraphs of history section.North8000 (talk) 18:16, 14 March 2022 (UTC)[reply]
  • Complete through history section including revolution subsection.North8000 (talk) 21:04, 15 March 2022 (UTC)[reply]
  • Complete through the Neural Networks section including all subsections. North8000 (talk) 14:23, 17 March 2022 (UTC)[reply]
  • Complete through hardware section. North8000 (talk) 00:07, 18 March 2022 (UTC)[reply]
  • Completed the articleNorth8000 (talk) 02:54, 18 March 2022 (UTC)[reply]

References[edit]

References

  1. ^ Gauen, Kent; Rangan, Rohit; Mohan, Anup; Lu, Yung-Hsiang; Liu, Wei; Berg, Alexander C. (2017-01). "Low-power image recognition challenge". 2017 22nd Asia and South Pacific Design Automation Conference (ASP-DAC). Chiba: IEEE: 99–104. doi:10.1109/ASPDAC.2017.7858303. ISBN 978-1-5090-1558-0. {{cite journal}}: Check date values in: |date= (help)
  2. ^ "Estimation of energy consumption in machine learning". Journal of Parallel and Distributed Computing. 134: 75–88. 2019-12-01. doi:10.1016/j.jpdc.2019.07.007. ISSN 0743-7315.

Natural language and processing out date[edit]

Now transformer base model out perform sequence model like lstm or gru and Google use Transformer base model to translate language 2402:800:6118:555E:10BE:9E4:1CEC:3EDE (talk) 18:15, 28 March 2022 (UTC)[reply]

Wiki Education assignment: Research Process and Methodology - FA23 - Sect 202 - Thu[edit]

This article was the subject of a Wiki Education Foundation-supported course assignment, between 6 September 2023 and 14 December 2023. Further details are available on the course page. Student editor(s): HELLOEXTRACREDIT (article contribs).

— Assignment last updated by HELLOEXTRACREDIT (talk) 04:30, 26 November 2023 (UTC)[reply]

Retrieved from "https://en.wikipedia.org/w/index.php?title=Talk:Deep_learning&oldid=1206847191"