Re:OpenAI O3 breakthrough high score on ARC-AGI-PUB
by torginus on Saturday December 21, 2024 @03:43AM
(#42478300)
Not sure how much that matters - I'm not an AI expert, but I did some intro courses where we had to train a classifier to recognize digits. How it worked basically was that we fed each pixel of the 2d grid of the image into an input of the network, essentially flattening it in a similar fashion.
It worked just fine, and that was a tiny network.
[ Parent
| on HN ]
Re:OpenAI O3 breakthrough high score on ARC-AGI-PUB
by thegeomaster on Saturday December 21, 2024 @10:04AM
(#42480079)
The classifier was likely a convolutional network, so the assumption of the image being a 2D grid was baked into the architecture itself - it didn't have to be represented via the shape of the input for the network to use it.
Re:OpenAI O3 breakthrough high score on ARC-AGI-PUB
by torginus on Saturday December 21, 2024 @11:41AM
(#42480601)
I don't think so - convolutional neural networks also operate over 1D flat vectors - the spatial relationship of pixels is only learned from the training data.
Re:OpenAI O3 breakthrough high score on ARC-AGI-PUB
by thegeomaster on Monday December 23, 2024 @07:20PM
(#42498647)
This is not true. CNNs perform 2D convolution, conceptually "sliding" a 2 dimensional kernel with learnable weights over the input image across two dimensions.
Perhaps it wasn't a convolutional network after all, but a simple fully-connected feed-forward network taking all pixels as input? Could be viable for a toy example (MNIST).