We’re introducing Jukebox, a neural net that generates music, including rudimentary singing, as raw audio in a variety of genres and artist styles. We’re releasing the model weights and code, along with a tool to explore the generated samples.

The TensorFlow blog contains regular news from the TensorFlow team and the community, with articles on Python, TensorFlow.js, TF Lite, TFX, and more.

The Gradient Boosting Machine is a powerful ensemble machine learning algorithm that uses decision trees. Boosting is a general ensemble technique that involves sequentially adding models to the ensemble where subsequent models correct the performance of…

Posted by Keith Hall, Research Scientist, Natural Language Understanding, Google Research Update — 2021/05/20: We are expanding the Research Expl...

The number of input variables or features for a dataset is referred to as its dimensionality. Dimensionality reduction refers to techniques that reduce the number of input variables in a dataset. More input features often make a predictive modeling task more…

Posted by Yuxiang Yang and Deepali Jain, AI Residents, Robotics at Google Recent advancements in deep reinforcement learning (deep RL) h...

Riemannian Adaptive Optimization Methods with pytorch optim - geoopt/geoopt

Project page for 'Scan-based Semantic Segmentation of LiDAR Point Clouds' with links to code and paper.

Video for "Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics" by A. Kuznetsov, P. Shvechikov, A. Grishin, D. Vetrov
Project page: https://bayesgroup.github.io/tqc
Paper: https://arxiv.org/abs/2005.04269
Code…

Datasets may have missing values, and this can cause problems for many machine learning algorithms. As such, it is good practice to identify and replace missing values for each column in your input data prior to modeling your prediction task. This is called…

A media server provides multimedia all-in-one features, such as video capture, processing, streaming, recording, and, in some cases, the ability to trigger actions under certain events, for example…

Text-to-speech systems across the industry typically rely on GPUs or specialized hardware to generate state-of-the-art speech in real-time production. We solved core efficiency challenges to process one second of audio in 500 milliseconds — using only CPUs.

Machine learning algorithms like Linear Regression and Gaussian Naive Bayes assume the numerical variables have a Gaussian probability distribution. Your data may not have a Gaussian distribution and instead may have a Gaussian-like distribution (e.g. nearly…