Artificial intelligence from Google goes public to crowdsource dev expertise
Google has launched an open source machine learning platform called TensorFlow, making it available to any developer who wants to tinker with the highly scalable system.
TensorFlow can already be found in the company’s speech recognition systems found on Google Now to its Photos photo search algorithms.
The web giant said TensorFlow can run on many devices, ranging from a single smartphone to an entire data centre.
“We’ve seen firsthand what TensorFlow can do, and we think it could make an even bigger impact outside Google,” wrote CEO Sundar Pichai on the official Google blog, explaining TensorFlow is faster, smarter, and more flexible than Google’s old system.
Machine-learning uses processes akin to how our brains work, with algorithms creating neural pathways to remember information that can easily be pulled up again. TensorFlow lets Google build and train “neural nets” up to five times faster than the old system.
“So today we’re also open-sourcing TensorFlow. We hope this will let the machine learning community—everyone from academic researchers, to engineers, to hobbyists—exchange ideas much more quickly, through working code rather than just research papers.”
“Although initially a research project, we’ve since collaborated with about 50 different teams at Google and deployed these systems in real products across a real wide spectrum of areas,” said Jeff Dean, a Google Senior Fellow.
“Today it’s used heavily in our speech recognition, in the new photos product, in Gmail, in Search.”