Convolutions vs Transformers
The cluster centers on debates about whether transformers or new ML architectures are essentially rediscovering or equivalent to convolutional neural networks (CNNs) and convolutions.
Activity Over Time
Top Contributors
Keywords
Sample Comments
Are Transformers based on convolutions?
Could this be used for convolutional neural networks?
I see we are re-deriving conv nets
But isn't this basically what the conv layer does...?
Convolutional neural networks are pretty big
Pretty broad but, from one section in the article, sounds like they're using convolutional layers in their networks?
It's just as bad a "convolutional neural networks" instead of "images being scaled down"
Convolution is fancy linear regression
They work at the level of convolutions, not images.
What's wrong about using the term convolution in ML?