Machine Learning

1 readers

1 users here now

Community Rules:

Be nice. No offensive behavior, insults or attacks: we encourage a diverse community in which members feel safe and have a voice.
Make your post clear and comprehensive: posts that lack insight or effort will be removed. (ex: questions which are easily googled)
Beginner or career related questions go elsewhere. This community is focused in discussion of research and new projects that advance the state-of-the-art.
Limit self-promotion. Comments and posts should be first and foremost about topics of interest to ML observers and practitioners. Limited self-promotion is tolerated, but the sub is not here as merely a source for free advertisement. Such posts will be removed at the discretion of the mods.

founded 1 year ago

MODERATORS

communick@academy.garden

[Discussion] Help needed with self supervised contrastive learning (alien.top)

submitted 11 months ago by SandMan1320@alien.top to c/machinelearning@academy.garden

2 comments fedilink hide all child comments

Hello,

So, i am working on a bit of a complex problem where I want to detect faults of a system using sensors data including cameras.

Since i only need two states pass / fail as output, I directed my attention to contrastive learning (specificlly self-supervised) version of it.

I am currently facing a challenge in defining the positive and negative pairs for training. In principle, I should use the image and its augmented versions as +ve pairs and the image and all other images as -ve. However, this is problematic as the other images may not be dissimilar.

So, any suggestions on how to tackle this problem ?

Thanks

top 2 comments

sorted by: hot top controversial new old

[–] OrganicCriticism6232@alien.top 1 points 11 months ago (1 children)

Don't think it matters. If there is any difference the encoder will be tuned to it.

[–] SandMan1320@alien.top 1 points 11 months ago

You mean that even if the pairs are inaccurate.. the network will handle that during tunning ?