Google’s Project Magenta: Blends Computing, Art And Music Using Artificial Intelligence

Humans are rather proud of their brains, and often use it to distinguish between one another. The human brain is one which is very distinct and bears unique properties allowing it to formulate mathematics, science, technology, the arts, literature and everything that they can intercept and interpret. In fact, they have proved that they could design a humanoid creature with an artificial brain, muscle, heart and everything under the sun which he requires to live, and probably make it even more perfect and sophisticated.

During Moogfest 2016, top researchers from Google had pronounced their initiative of making an open source project with the help of Artificial Intelligence and deep learning process. What was in their mind? The research objective is pretty simple, yet illusionary and would take thousands of neurons to work in sync to produce such an output. Douglas Eck, in his speech articulated that the “Google Brain”, the core section which works behind the scene to engineer the world has engaged themselves in the making of an AI based system which could weave its imagination on canvas.


paintings_of_magneta
Magenta's Own Art

We can think of robots lifting heavy objects or taking care of people or even playing drums. But something like painting, where a direct conjugation of an abstract entity like imagination is required to sync with the medium is much more complex than it sounds. With such an enthralling vision in its mind, Google is all set to officially launch the project “Magenta” which is based on the open source Software library TensorFlow. The software is available on GitHub and as a public forkable repository. It’s basically a machine learning tool that can perform heavy computation based on data flow graphs. Also, inspired by the DEEPDREAM tool, the deep core of artificial neural networks was realized, which in turn provided a detailed understanding of intuition and imagination and how an algorithm could perceive an object in his way.

Douglas said that in the future, they like to move to music and videos and see how their project can actually react to that. Although initiated by Google, the research team have adapted to open sourcing and shared the endeavour with all. Currently the GitHub project page does not have much except a “read me” .md file but soon it would be filled with prospective commits. In spite of having a serious goal, the project, in background wants to blend computation with visual and vocal arts and thereby create a larger community.

Source: #-Link-Snipped-# | #-Link-Snipped-# |#-Link-Snipped-#

Replies

You are reading an archived discussion.

Related Posts

A mechanical engineer from IIT Mandi, Saksham Grover decided to build 'DriversKart' when he realised that commuting options should not be limited to hiring cabs or autos. There are times...
Quote: A few weeks ago, I finally tried Tesla Motors' "autopilot" feature. A Tesla rep and I tooled around Houston's I-45 in a Model X crossover SUV for 15 minutes,...
The numerous accounts you have accross the web and the "one-account,access-all" vital Gmail, the core control element might be out of passwords in the literal sense. Google has something up...
Quote: International Journal of Engineering Research and Technology is peer-reviewed, online international journal published monthly by ESRSA Publication. IJERT is a highly-selective journal, covering topics that appeal to a broad...
German Automobile giant, Volkswagen has recently unleashed its new sub 4-meter length Sedan Ameo at the Chakan Plant in Pune, India. Aimed to enthrall adventure enthusiasts, this car could be...