An in-depth explanation of challenges and one potential remedy for gradient estimation in stochastic computational graphs youtu.be/nKM9875PVtU
Variational Autoencoder, Latent Variable Model & Amortized Inference... youtu.be/h9kWaQQloPk
What is Probabilistic Programming? + A quick & opinionated review of @mcmc_stan @pymc_devs @PyroAi @TuringLang @TensorFlow Probability (TFP) youtu.be/-C8PGgnF1yg using an example from @rlmcelreath's Statistical Rethinking book
The Metropolis Algorithm. Watch the full tutorial (The Metropolis-Hastings Algorithm) on youtube (youtube.com/watch?v=oX2wIGSn…)
A visual introduction to Rejection Sampling! Full tutorial on YouTube youtu.be/si76S7QqxTU
Published a video explaining "MADE: Masked Autoencoder for Distribution Estimation" youtu.be/7q4ueFiJjAY MADE is an essential component in many modern Normalizing Flows algorithms
Published a comprehensive tutorial on Normalizing Flows - Motivations, The Big Idea, & Essential Foundations youtu.be/IuXU2dBOJyw
ksachdeva retweeted
But what is Entropy? by @ksachdeva buff.ly/2DxeHYK
ksachdeva retweeted
If you are hearing about #covid19 “reinfections” in Asia, I can offer you my take as a virologist. The best explanation for what we’re seeing is likely due to three things...
Statistical Rethinking (2nd Ed) Port to Tensorflow Probability github.com/ksachdeva/rethink… Thanks @rlmcelreath @fehiepsi @junpenglao
Statistical Rethinking - Feeling you get after reading this book - "I am not an idiot after all, it was just that I had not found the right book !" Much respect & gratitude professor @rlmcelreath
A Hitchhiker's Guide to designing a Bayesian library in Pyt... youtu.be/WHoS1ETYFrw Many Thanks @junpenglao this one is a gem and came my way at just the appropriate time.
ksachdeva retweeted
A unified method for downloading, extracting, processing & using datasets by @ksachdeva buff.ly/34OAynH
ksachdeva retweeted
Wow, this is great
Machine Learning, Online Comic from @GoogleAI
cloud.google.com/products/ai…
ksachdeva retweeted
New blog post on vectorizing MCMC, and pointing out that TensorFlow Probability makes running hundreds of chains practically free.
I include a numpy implementation so I could build this demonstration using 4,096,000 samples from a 2D Gaussian:
colindcarroll.com/2019/08/18…
ksachdeva retweeted
As always, reading @aubreyclayton articles is a pleasure! Great overview of the relationship between prior probabilities and the replication crisis!
"Atheists hailed it as scientific proof that religion was irrational; religious people were understandably offended at the suggestion that the source of their faith was a lack of reasoning ability." bit.ly/flawed-reason
Excellent work by @justadudewhohax (I like to call him #awesomedudewhohax) xitter.cf/justadudewhohax/status…
Finally #tensorflowjs implementation of MTCNN for Realtime Face Detection and 5 Point Face Landmark Detection is available in face-api.js v0.8.0. Check out the example for live face tracking from your webcam! #javascript #facedetection #tensorflow #js github.com/justadudewhohacks…












