603
OpenAI is reportedly going all-in as a for-profit company (mashable.com)
submitted 2 years ago* by to c/technology@lemmy.world
119 comments fedilink hide all child comments
 

Surprised pikachu face

you are viewing a single comment's thread
view the rest of the comments
[–] 5 points 2 years ago (2 children)

Other than citing the entire training data set, how would this be possible?

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 2 years ago (1 child)

    The entire training set isn't used in each permutation. Your keywords are building the samples based on metadata tags tied back to the original images.

    If you ask for "Iron Man in a cowboy hat", the toolset will reach for some catalog of Iron Man images and some catalog of cowboy hat images and some catalog of person-in-cowboy-hat images, when looking for a basis of comparison as it renders the image.

    These would be the images attributed to the output.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •
  • [–] 2 points 2 years ago

    Do you have a source for this? This sounds like fine-tuning a model, which doesn't prevent data from the original training set from influencing the output. The method you described would only work if the AI is trained from scratch on only images of iron man and cowboy hats. And I don't think that's how any of these models work.

  • permalink
  • fedilink
  • source
  • parent