ICML 2024 Tutorial"Machine Learning on Function spaces

[@abhijeetvyas7365]

Insightful talk - 37:31 is classic for Fourier basis.

[@Yuan-HengWang]

Excellent, thanks for sharing! Adding new learning task into my list.

[@galoisgogh452]

Great talk, but I think the claim of delta xj can be cancelled out was made in haste. Given that you have two netted integration/summation, the using of same variable name for inner integration/summation and the outer summation/integration created confusion, perhaps once you put a hat/tilde in one integration/summation, it then became clear. Am I missing something?

[@picpic-k3c]

insightful talk! 👍👍

[@jesusmtz29]

Do you have a ref to a proof of the statement at 21:32? Great talk

[@kamyarazizzadenesheli7740]

The concept is defined and proven in this work . Look at section 2.3. arxiv.org/pdf/2108.08481.pdf

[@themosst]

Thanks for the video. At 30:00 you switch from a(x) to a(y) in the integral. Is this an error?

[@kamyarazizzadenesheli7740]

Good catch. Mathematically it is correct, but it would have been more appealing to have a(x) instead of a(y), as you mentioned, then, we should have had \int \kappa(y,x)a(x)dx + ... .

[@TG-bf8um]

Is there a version of this tutorial where the video slides are visible?

[@kamyarazizzadenesheli7740]

The links are on the discerption above.

[@mojtabakolahdoozi2418]

So insightful talk!

[@alirezamogharabi8733]

آقا ایرانی هستی؟ دمت گرم باعث افتخار ما هستید شما

[@babakadelinasab]

بله ایرانیه.بچه آمل.پسر دایی بنده

[@babakadelinasab]

بله ایرانی هست ولی هیچکی نمیشناستش واین واقعا بده

[@babakadelinasab]

کامیار کی بودی تو؟