Users appreciate the ICML tutorial's honest discussion of gaps between optimization theory and real-world ML practice, especially around Adam Optimizer.
Based on 2 visible X reactions from 5 accounts; directional sample.
Ask a question below.
Published answers will appear here.
Empirical optimizers like Adam dominate despite lacking theoretical proofs.
My brain is pretty split between ML I know how to teach and ML I do in practice. Really appreciate the honesty about when these are bridgeable.
@chrmanning @MarkSchmidtUBC @CIFAR_News Quiet experts are the ones actually building the foundations most people stand on.
Fantastic slides from the ICML tutorial of @MarkSchmidtUBC (“is opt theory relevant in 2026”) https://www.cs.ubc.ca/~schmidtm/Documents/2026_ICML_Tutorial.pdf
I remember @MarkSchmidtUBC very fondly from @CIFAR_News meetings as a quiet expert. Science has the egos and the egoless.
Excellent tutorial, a must see
Users appreciate the ICML tutorial's honest discussion of gaps between optimization theory and real-world ML practice, especially around Adam Optimizer.
Based on 2 visible X reactions from 5 accounts; directional sample.
Ask a question below.
Published answers will appear here.