"Bayesian Data Science by Simulation" #PyCon2019
β ML Review (@ml_review) May 9, 2019
3 hour video introduction to Bayesian data science through the lens of simulation By @ericmjl @hugobownehttps://t.co/RsmwPWK5b0
"Bayesian Data Science by Simulation" #PyCon2019
β ML Review (@ml_review) May 9, 2019
3 hour video introduction to Bayesian data science through the lens of simulation By @ericmjl @hugobownehttps://t.co/RsmwPWK5b0
From Variational to Deterministic Autoencoders
β hardmaru (@hardmaru) April 10, 2019
They show that simple regularized autoencoders can also learn a smooth, meaningful latent space without having to force the bottleneck code to conform to some arbitrarily chosen prior (ie Gaussian for VAEs).https://t.co/rsmiXk8Vw6
Some Stan and Bayes short courses! https://t.co/LW9FN45j7V
β Andrew Gelman (@StatModeling) April 4, 2019
TF Probability now supports Keras! Check out their talk, you'll need details on this one -- but wow! pic.twitter.com/PgNVdoOeVn
β Josh Gordon (@random_forests) March 7, 2019
π― βDespite its prevalence, censored time-to-event data is often overlooked, leading to dramatically biased predictions.β https://t.co/C8uvtPC6OM
β Erik Bernhardsson (@fulhack) March 2, 2019
π€£ This is hilarious β the list of prior beliefs about Bayesian stats is GOLD!
β Mara Averick (@dataandme) February 19, 2019
"Duncan Garmonsway: Rebel Bayes Day 1" by @nacnudushttps://t.co/lALXwBRTW9 (also feat. useful learnings from @rlmcelreath's Statistical Rethinking)
New work with the Pyro team (https://t.co/6en944fTtd) on efficient batched discrete variable inference through tensor variable elimination.
β harvardnlp (@harvardnlp) February 13, 2019
Practically generalizes 'einsum' with batching, semirings (log-space, max, etc), and marginals. Powers Pyro's discrete directed inference. pic.twitter.com/upE6Soazkw
π This is a great, quick tour through diff. statistical methods:
β Mara Averick (@dataandme) January 30, 2019
"Half a dozen frequentist and Bayesian ways to measure the difference in means in two groups" π¨βπ» @andrewheiss https://t.co/6BcZ4HK7WW #rstats #rstan pic.twitter.com/7xO9KHRj3Q
2/2
β Judea Pearl (@yudapearl) January 30, 2019
a narrower view of Bayesianism: "if you don't assign priors to parameters you ain't a Bayesian." I dont buy it, do you? Priors on parameters is just one way of encoding prior knowledge, and not a very good one.
Have you ever wanted to see seven different Bayesian and frequentist ways to measure the avg difference between two groups? Check out my new post to see examples of standard #rstats t-tests, the infer package, brms regression, and raw @mcmc_stan code! https://t.co/U2sjL9nt6b pic.twitter.com/OgK1zsQXiY
β Andrew Heiss, PhD (@andrewheiss) January 29, 2019
π€ R, Bayes and #dataviz? @mjskay brings the magic (always, and in this thread)! β¨ #rstats https://t.co/6FiU6mo7aT
β Mara Averick (@dataandme) January 21, 2019
Interested in statistical methods for data science? Check out our probabilistic programming and bayesian methods playlist, including:
β PyData (@PyData) January 20, 2019
"Bayesian Network Modeling using R and Python" by Pragyansmita Nayak: https://t.co/CNw92p8jAJ
Don't forget to subscribe to #PyData on Youtube!