Jump to content

Talk:Neuromorphic computing: Difference between revisions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Browse history interactively
← Previous editNext edit →
Content deleted Content added
VisualWikitext
Assessment: Engineering: class=B; Neuroscience: class=B (assisted)
No edit summary
Line 1: Line 1:
{{WikiProject Engineering |class=B}}
{{WikiProject Engineering |class=B}}
{{WikiProject Neuroscience |class=B |importance=low}}
{{WikiProject Neuroscience |class=B |importance=low}}
{{WikiProject Computing
|class=
|importance=
|hardware=yes
|hardware-importance=
}}
{{Connected contributor (paid)|User1=Inkian Jason|U1-employer=Beutler Ink|U1-client=[[Intel]] via Interfuse Communications}}
{{Connected contributor (paid)|User1=Inkian Jason|U1-employer=Beutler Ink|U1-client=[[Intel]] via Interfuse Communications}}
{{findsourcesnotice|neuromorphic (engineering OR computing)}}
{{findsourcesnotice|neuromorphic (engineering OR computing)}}

Revision as of 11:03, 26 June 2020

WikiProject iconEngineering B‑class
WikiProject iconThis article is within the scope of WikiProject Engineering, a collaborative effort to improve the coverage of engineering on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.EngineeringWikipedia:WikiProject EngineeringTemplate:WikiProject EngineeringEngineering
BThis article has been rated as B-class on Wikipedia's content assessment scale.
???This article has not yet received a rating on the project's importance scale.
WikiProject iconNeuroscience B‑class Low‑importance
WikiProject iconThis article is within the scope of WikiProject Neuroscience, a collaborative effort to improve the coverage of Neuroscience on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.NeuroscienceWikipedia:WikiProject NeuroscienceTemplate:WikiProject Neuroscienceneuroscience
BThis article has been rated as B-class on Wikipedia's content assessment scale.
LowThis article has been rated as Low-importance on the project's importance scale.
WikiProject iconComputing Unassessed
WikiProject iconThis article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing
???This article has not yet received a rating on Wikipedia's content assessment scale.
???This article has not yet received a rating on the project's importance scale.
Taskforce icon
This article is supported by Computer hardware task force.

Template:Findsourcesnotice

This mass of external links was removed from the article per the WP:EL guidelines. The list has been copied here for those who are interested but please do not restore them to the article.


Below is a list of leading researchers in various areas of neuromorphic engineering:

Vision and Robotics

Hearing, Sonar, and Speech Processing

Neural Processing and Processing and Learning

Spike-Based Processing

Biomedical and Ultra-Low-Power Applications


(See Leading Labs)

Other Links

Assessment

I'm responding to a request for reassessment from WP:WikiProject Neuroscience. I left it at Start Class, and rated it as Low Importance. Given the request, I'll comment on my reactions to the page. I think that it suffers from WP:NOTDICT and WP:SYNTH. It seems to me to start out as a dictionary-like definition of a term that is not particularly notable in the neuroscience literature, and then patches together a series of examples, without a clear indication that secondary sources regard these examples as being related by their belonging to this term. The most notable example given is the Human Brain Project, but it is unclear whether characterizing the project as being neuromorphic engineering is merely original research. Sorry for being harsh, but that's the way I see it. I hope that other editors will be able to improve the page. Alternatively, perhaps it should simply be merged into Computational neuroscience. --Tryptofish (talk) 22:17, 12 March 2013 (UTC)[reply]

Merge with Cognitive Computing

It has been suggested that Cognitive computing be merged into this article. (Discuss) Proposed since September 2014.

The article now appears with the banner above. I would like to oppose this change. If anything, the merge would be the other direction (as cognitive computing includes both neuromorphic computing and specific kinds of AI processing such as that done in the Watson system), but I would also oppose that. The two are now being treated in the research community as related topics with different uses - for example the Journal of Cognitive Computing will publish, but is not limited to, neuromorphic computing papers. The neural network literature does not publish papers on the more symbolic approaches, but do publish papers on using neural approaches to solve similar problems. I would thus suggest leaving these as separate articles that point at each other. JAHendler (talk) 00:31, 12 December 2014 (UTC)[reply]

Both the proposed merge to Cognitive computing and Physical neural network is equally weird. The first is about computing while the other is about physical engineering. Neuromorphic engineering (I prefer neuromorphic computing) is about how to do actual processing in available materials the way the brain does it. One can say that neuromorphic engineering/computing connects cognitive computing with physical engineering. Note that neural networks is a specific type of simplified early models for how the brain was supposed to work, and is not how the brain actually works. ;) Jeblad (talk) 18:01, 17 April 2016 (UTC)[reply]
I agree that the Cognitive computing merge doesn't make sense, since that article is talking about software algorithms for human–computer interaction, while this one is about hardware architectures that mimick neural networks. I've removed that merge tag. I'm inclined to support the merge from Physical neural network, though, since the subject matter of the articles seems identical. Antony–22 (talk⁄contribs) 19:16, 18 April 2016 (UTC)[reply]
Sorry, but Cognitive computing isn't about software algorithms for human–computer interaction. Let this be until someone that knows the field has time to look into it, I don't have the time right now. Jeblad (talk) 10:11, 20 April 2016 (UTC)[reply]
Well, I was going off of the definition in the lead of that article, which seems to have changed in the last two days. Now it's so vague I don't know what it is.
Upon further thought, physical neural networks are just one way to do neuromorphic computing, so perhaps they should stay as separate articles. But as it stands, the content of both articles is mainly lists of implementations of neuromorphic analog electronics, which should moved to either one article or the other. Antony–22 (talk⁄contribs) 17:37, 20 April 2016 (UTC)[reply]

Stochastic phase-change neurons

IBM Research in Zurich has created the world's first artificial nanoscale stochastic phase-change neurons.

Artificial neuromorphic systems based on populations of spiking neurons are an indispensable tool in understanding the human brain and in constructing neuromimetic computational systems.

Article URL: http://arstechnica.com/gadgets/2016/08/ibm-phase-change-neurons/

Study: http://www.nature.com/nnano/journal/v11/n8/full/nnano.2016.70.html — Preceding unsigned comment added by Kenorb (talk • contribs) 00:56, 6 August 2016 (UTC)[reply]

Neuromorphic

It could be worth nothing that some describe all kinds of low-power neuro-lookalike VLSI implementation as neuromorphic other believe that only implementations that utilize ideas from biological neurons should be said to be neuromporphic. In particular; an implementation that uses backprop algorithms is not based upon biological neurons as those use other means to reinforce learning at the synaptic level. Ie. backprop would imply learning at the neuron level, and in particular information flow upward along the axon. Jeblad (talk) 15:18, 21 April 2017 (UTC)[reply]

Hello fellow Wikipedians,

I have just modified one external link on Neuromorphic engineering. 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:

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) 14:49, 16 February 2018 (UTC)[reply]

Request to move mention of Intel Loihi from "See also" to "Examples"

Hello! I'd like to propose an update to this Wikipedia article. But first, a disclosure, which is also displayed in the connected contributor template above: I am submitting this request on behalf of Intel via Interfuse Communications, and as part of my work at Beutler Ink. Given my conflict of interest, I'm seeking volunteer editors to review and implement the proposed content appropriately, and I will not be editing the article directly.

Currently, the article's "Examples" section mentions work by several companies and institutions, including the Georgia Institute of Technology, MIT, Purdue, HP Labs, and IBM. "Intel Loihi" is currently displayed as a link within the "See also" section, but given the chip's significance and mentions in the AI accelerator and Cognitive computing entries, I believe a longer mention would provide useful information not currently found in the entry.

I propose removing the "See also" link and adding the following sentence to the "Examples" section:

Intel unveiled its neuromorphic research chip, called "Loihi", in October 2017. The chip uses an asynchronous spiking neural network (SNN) to implement adaptive self-modifying event-driven fine-grained parallel computations used to implement learning and inference with high efficiency.[1][2]

References

  1. ^ Davies, Mike; et al. (January 16, 2018). "Loihi: A Neuromorphic Manycore Processor with On-Chip Learning". IEEE Micro. 38 (1): 82–99. Retrieved August 5, 2018.
  2. ^ Hsu, Jeremy (January 9, 2018). "CES 2018: Intel's 49-Qubit Chip Shoots for Quantum Supremacy". Institute of Electrical and Electronics Engineers. Retrieved August 5, 2018.

This is a short description of the chip, which I believe is more beneficial to the article than a redirect in the "See also" section. I'd suggest placing this content at the end of the "Examples" section, since the last three paragraphs don't really have specific dates, but editors may prefer somewhere else to maintain chronology. Do other editors think this is an appropriate addition to the article? If so, feel free to copy and paste the following markup for easier implementation in the main space:

Markup

[[Intel]] unveiled its neuromorphic research chip, called "[[Intel Loihi|Loihi]]", in October 2017. The chip uses an asynchronous [[spiking neural network]] (SNN) to implement adaptive self-modifying event-driven fine-grained parallel computations used to implement learning and inference with high efficiency.<ref>{{cite journal |last1=Davies |first1=Mike |title=Loihi: A Neuromorphic Manycore Processor with On-Chip Learning |journal=IEEE Micro |date=January 16, 2018 |volume=38 |issue=1 |pages=82–99 |url=https://ieeexplore.ieee.org/document/8259423/ |accessdate=August 5, 2018 |display-authors=etal}}</ref><ref>{{cite web |last1=Hsu |first1=Jeremy |title=CES 2018: Intel's 49-Qubit Chip Shoots for Quantum Supremacy |url=https://spectrum.ieee.org/tech-talk/computing/hardware/intels-49qubit-chip-aims-for-quantum-supremacy |publisher=[[Institute of Electrical and Electronics Engineers]] |accessdate=August 5, 2018 |date=January 9, 2018}}</ref>

Let me know if you have any questions or concern, and thanks for your consideration. Inkian Jason (talk) 21:37, 6 August 2018 (UTC)[reply]

There is secondary coverage of this chip by now, at least in the popular press, so I would be comfortable adding a short description to the examples section. The second reference above is about quantum computers and just mentions Loihi in passing. Perhaps the ZDNet article would be a better alternative? It goes more in depth and compares it with other efforts like True North. --{{u|Mark viking}} {Talk} 19:16, 17 August 2018 (UTC)[reply]
@Mark viking: Thanks for reviewing. I am fine using your preferred sourcing. I don't edit the main space directly because of my COI. Do you mind updating the article appropriately? Inkian Jason (talk) 19:37, 17 August 2018 (UTC)[reply]
 Done --{{u|Mark viking}} {Talk} 23:28, 17 August 2018 (UTC)[reply]
@Mark viking: Thanks for your help. I noticed the text "hideMarkup" now appears in the article's prose, but otherwise the addition looks good. Inkian Jason (talk) 14:19, 20 August 2018 (UTC)[reply]
Fixed, thanks. --{{u|Mark viking}} {Talk} 17:40, 20 August 2018 (UTC)[reply]
@Mark viking: Thanks again! Inkian Jason (talk) 17:41, 20 August 2018 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Talk:Neuromorphic_computing&oldid=964584955"