Pinned post
Here are the slides from my talk at #NIPS2016 on machine learning & likelihood-free inference for particle physics
dx.doi.org/10.6084/m9.fig…
dx.doi.org/10.6084/m9.fig…
Kyle Cranmer
Verified account This account is verified. Learn more@KyleCranmer
Director Data Science Institute @UWMadison @datascience_uw. EiC @MLSTjournal.
Physics, stats/ML/AI, open science.
same handle @sigmoid.social and bsky
Physics, stats/ML/AI, open science.
same handle @sigmoid.social and bsky
3.4K Following 18K Followers 28K Posts
Latest posts 5
This is the way
One of my blackboards from this semester’s Data Sciences course: explaining the asymptotic consistency of the maximum likelihood estimator. I’m teaching it in English this year—should I put the whole course on YouTube? :)View quoted post on X
Last year, I had a conversation with @RichardSSutton on the Bitter Lesson and wrote an essay entitled "The Bittersweet Lesson". It's not a counter to the original, but a statement about how our confusion on this topic points to issues to the need for deeper understanding.
theoryandpractice.org/2025/09/The%20…
theoryandpractice.org/2025/09/The%20…
theoryandpractice.org The Bittersweet Lesson In 2019, Rich Sutton authored an essay entitled “The Bitter Lesson”, where he made the observation that general purpose methods that leverage computation eventually win over bespoke solutions that... Brute force computation is not enough for intelligence. Clever new methods are needed. The science of AI is far from finished. IMHOView quoted post on X
Whacky that there is such a large, finite object that has such an simple, infinite, and elegant property
hello there ℤ is Diophantine in ℚ, thanx to my close friends Astra and Claude
The polynomial characterizing ℤ isn’t quite Tweet-sized; it has m=292,890,231,905 variables, is of degree d < 2^^7, and all its coefficients have absolute value < 2^^8.View quoted post on X
