Artificial Intelligence

Posit AI Blog: Getting started with Keras from R

If you’ve been thinking about diving into deep learning for a while – using R, preferentially –, now is a good time. For...

Gaussian Process Regression with tfprobability

How do you motivate, or come up with a story around Gaussian Process Regression on a blog primarily dedicated to deep learning? Easy. As...

R interface to TensorFlow Hub

We are pleased to announce that the first version of tfhub is now on CRAN. tfhub is an R interface to TensorFlow Hub...

Posit AI Blog: Differential Privacy with TensorFlow

What could be treacherous about summary statistics? The famous cat overweight study (X. et al., 2019) showed that as of May 1st, 2019, 32...

First experiments with TensorFlow mixed-precision training

Starting from its - very - recent 2.1 release, TensorFlow supports what is called mixed-precision training (in the following: MPT) for Keras. In...

NumPy-style broadcasting for R TensorFlow users

We develop, train, and deploy TensorFlow models from R. But that doesn’t mean we don’t make use of documentation, blog posts, and examples...

Posit AI Blog: Infinite surprise

Among deep learning practitioners, Kullback-Leibler divergence (KL divergence) is perhaps best known for its role in training variational autoencoders (VAEs). To learn an...

Posit AI Blog: Introducing: The RStudio AI Blog

Why the new name, RStudio AI Blog? There is a straightforward reason. The previous title, “TensorFlow for R Blog”, was a good match for...

A first look at federated learning with TensorFlow

Here, stereotypically, is the process of applied deep learning: Gather/get data; iteratively train and evaluate; deploy. Repeat (or have it all automated as a continuous...

Latest articles