Thread

1. akkad33 · 1789979305 · focus · HN ↗

   Can someone tell me what is the difference between Jev and a normal neural network that does classification ?

   My understanding is: it takes text input and it does one shot classification (no training data)

   1. crackalamoo · 1789979441 · focus · HN ↗

      Yes, this is essentially it.

      As a corollary, the output classes can be any set, rather than needing to be set before training.

      1. akkad33 · 1789981358 · focus · HN ↗

         Can someone do a ELI5A of how they achieve classification over any user defined list of items? Normal neural networks do a softmax over a known output set to get probabilities

         1. andy12_ · 1789986346 · focus · HN ↗

            You can achieve open-vocabulary classification by making the final weights in the softmax come from a category encoder instead of being fixed learned weights. So instead of

            softmax(encode(input)*learned_weights)

            You have

            softmax(encode(input)*encode(categories))

            I'm not sure if Jev does it this way, but it's how you get open-vocabulary zero-shot image classification with models like CLIP [1].

            [1] <a href="https:&#x2F;&#x2F;openai.com&#x2F;index&#x2F;clip&#x2F;" rel="nofollow">https:&#x2F;&#x2F;openai.com&#x2F;index&#x2F;clip&#x2F;

            1. nighthawk454 · 1790049186 · focus · HN ↗

               which amounts to nearest-neighbor

Unofficial Hacker News client; not affiliated with Y Combinator.