L13.5 What's The Difference Between Cross-Correlation And Convolution?

Sebastian's books: sebastianrasch...
This replaces a previous video where the video & audio were out of sync.
Slides: sebastianrasch...
Code: github.com/ras...

[@homataha5626]

Can't wait for you to do a video on Diffusion Models !

[@SebastianRaschka]

Haha, same! So many other things on my todo-list that I don't want to abandon, but yeah, I hopefully get to it some day later this year :)

[@ChrisOffner]

3:48 I'm a bit confused by the index convention with negative indices. I'm assuming you use the {(-1, -1), ..., (1, 1)} indices only for the kernel K but not for the input A or output Z, but then A[i + u, j + v] would index into A with negative indices if u or v are negative (e.g. -k) and i and j are small. I think I'd prefer just normal matrix indexing across all matrices involved.

[@zigzag4273]

Hey Seb. Hope you're well, could you make a video of how you read and replicate papers.

[@SebastianRaschka]

Good suggestion. I meant to do something like that ... it's actually on my list. I just got too busy with some other content. But if I ever find the time, this would be a good one!

[@user-jp5ir6fc5f]

Please do that
@@SebastianRaschka

[@user-wo6tl6ci9s]

You the best

[@EducatedButton]

Can you please do a ChatGPT video similar to the ones you have for GPT and BERT models.