Re: [Wikimedia-l] Siamese networks and image classification

[Pine W]

Hi John,

I have not heard of an initiative to use Siamese neural networks for image
classifications on on Commons. You might make a suggestion on the AI,
Research, and/or Commons mailing lists regarding this idea. You might also
make a suggestion in IdeaLab
.

Pine 


On Sun, Jan 14, 2018 at 3:46 AM, John Erling Blad  wrote:

> Has anyone tried to use a Siamese neural network for image classification
> at Commons? I don't know if it will be good enough to run in autonomous
> mode, but it will probably be a huge help for those that do manual
> classification.
>
> Imagine a network providing a list of possible categories, and the user
> just ticks off usable categories.
>
> A Siamese network can be learned by using a triplet loss function, where
> the anchor and the positive candidate comes from the same category, and the
> negative candidate comes from an other category but are otherwise close to
> the anchor.
>
> Output from the network is like a fingerprint, and those fingerprints can
> be compared to other images with known fingerprints, or against a
> generalized fingerprint for a category.
>
> John Erling Blad
> ___
> Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/
> wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/
> wiki/Wikimedia-l
> New messages to: Wikimedia-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> 
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 

[Wikimedia-l] What's making you happy this week? (Week of 14 January 2018)

[Pine W]

I appreciate Legoktm's initiative for code review in 2018:
https://lists.wikimedia.org/pipermail/wikitech-l/2018-January/089381.html

What's making you happy this week?

Pine 

___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, 

[Wikimedia-l] Siamese networks and image classification

[John Erling Blad]

Has anyone tried to use a Siamese neural network for image classification
at Commons? I don't know if it will be good enough to run in autonomous
mode, but it will probably be a huge help for those that do manual
classification.

Imagine a network providing a list of possible categories, and the user
just ticks off usable categories.

A Siamese network can be learned by using a triplet loss function, where
the anchor and the positive candidate comes from the same category, and the
negative candidate comes from an other category but are otherwise close to
the anchor.

Output from the network is like a fingerprint, and those fingerprints can
be compared to other images with known fingerprints, or against a
generalized fingerprint for a category.

John Erling Blad
___
Wikimedia-l mailing list, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: Wikimedia-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,