Google adds distributed computing to TensorFlow, its Open Source Machine Learning platform

Google just announced TensorFlow update with the much requested feature of running the machine learning tool on multiple computers

Google announced that its Open Source software library for Machine Intelligence now supports distributed computing. TensorFlow was developed by Google Brain to cater to the demands of various Google Applications that use Machine Learning. the process of Machine Learning, has a stage known as "learning" which is demonstrated in this browser app, a Neural Network Playground that also uses TensorFlow.

Distributed computing allows neural networks to learn much faster than the network running on one computer. This was one of the most requested features by developers and allows tens or hundreds of computers to work in parallel, and reduces the time taken for some tasks from weeks to hours.

The Verge has a report covering some of the more compelling projects that developers have created using TensorFlow. These include a computer that learns to play pong, and a neural network that invents fake new Chinese characters. These are the kinds of projects that stand to benefit by the distributed computing update.

Business Insider describes the excitement that machine learning enthusiasts have for this new release, with a headline that says "Programmers are going crazy for free Google software that creates self-learning computers". TensorFlow is one of the most accessible and easy ways to set up a neural network.

The Guardian has a more in depth piece outlining the challenges of machine learning, and why tasks that are trivially easy for humans can be prohibitively difficult for computers.

Find our entire collection of stories, in-depth analysis, live updates, videos & more on Chandrayaan 2 Moon Mission on our dedicated #Chandrayaan2TheMoon domain.