Fascinating intellectual "battle". Statistics is a field that should reinvent itself soon. Could this be the beginning? https://t.co/WEOvnZns2g
β Xavierππ€π (@xamat) January 9, 2019
Fascinating intellectual "battle". Statistics is a field that should reinvent itself soon. Could this be the beginning? https://t.co/WEOvnZns2g
β Xavierππ€π (@xamat) January 9, 2019
Statistical thought of the day: I don't embrace the Bayesian paradigm because it is without problems. I embrace it because (1) it solves the most problems and (2) null hypothesis significance testing, p-values, and type I error are beyond repair. Mixing Bayes+frequentist=mess.
β Frank Harrell (@f2harrell) January 8, 2019
OH WAIT HERE'S THE LINKhttps://t.co/I5D7LtDObP
β alex hayes (@alexpghayes) December 25, 2018
The International Conference on Probabilistic Programming
β ML Review (@ml_review) December 16, 2018
Talks from the #PROBPROG 2018 Conference, held at the MIT Media Lab in Cambridgehttps://t.co/kstWPLMiQF pic.twitter.com/yLM2uhhtY0
This is a great summary.
β Jeremy Howard (@jeremyphoward) December 14, 2018
Most people only mention points 3&6 (which are kinda the same thing?), and are the only ones I'd consider controversial (since any model can be trivially adjusted to output a distribution, so it may not be a big win) https://t.co/Vobu4WEyyO
Good Part 5: Model checking as a core activity
β Sean J. Taylor (@seanjtaylor) December 14, 2018
Good Bayesian analyses consider a wide range of models that vary in assumptions and flexibility in order to see how they affect substantive results. There are principled, practical procedures for doing this.
Good Part 2: No need to derive estimators
β Sean J. Taylor (@seanjtaylor) December 14, 2018
There are a increasingly full-featured and high-quality tools that allow you to fit almost any model you can write down. Being able to treat model fitting as an abstraction is great for analytical productivity.
A couple days ago another team asked me to speak about Bayesian data analysis. I decided that instead of doing a nuts/bolts of how to fit/use Bayesian models, I would describe "Bayesian analysis: The Good Parts". <potentially controversial thread>
β Sean J. Taylor (@seanjtaylor) December 14, 2018
Learn probabilistic programming with TensorFlow Probability, from the ground up. The Bayesian Methods for Hackers book is now available in open source in TFP!
β TensorFlow (@TensorFlow) December 10, 2018
Read post here β https://t.co/KrvlXM0lte
High-resolution PDF available here: https://t.co/5OmlGx91qR pic.twitter.com/tn3KLXZg1S
β Dustin Tran (@dustinvtran) December 7, 2018
Optuna β Bayesian hyperparameter optimisation framework with pruning and parallelisation. Features an imperative & modular define-by-run style API
β ML Review (@ml_review) December 4, 2018
By @iwiwi @toshihikoyanase https://t.co/Of3NO3Ounf pic.twitter.com/GKE9JrAds9
Making Your Neural Network Say βI Donβt KnowββββBayesian NNs using Pyro and PyTorch by @paraschopra https://t.co/uIJXniArqw #deeplearning #machinelearning #ml #ai #neuralnetworks #datascience #pytorch
β PyTorch Best Practices (@PyTorchPractice) November 30, 2018