Quantum Computers aren’t Ideal for Deep Learning

Quantum Computers aren’t Ideal for Deep Learning

Quantum computers aren’t ideal for deep learning said by a Google researcher
In the past years, Google has tried to get better more and more about its services with AI. Google also happens to have a quantum computer a system who is able to do certain computations quicker than classical computers.

Google is enthusiastic on advancing its capabilities in a type of AI known as deep learning, which includes training ANN (artificial neural networks) on a large provider of data and then getting them to make deductions about new data.

An event was held at Google headquarters last week; in which a Google researcher explained that the quantum computing t isn’t just the best fit for systems.

Several other technology companies including Facebook and Microsoft have tested with deep learning in the background of image recognition, natural processing, and speech recognition. Those other companies are huge, with abundance of money to spend on communications. But they don’t have quantum computer while Google does. Still, that doesn’t mean it’s always the top tool for the job.

Deep learning, although, is another thing. Normally speaking, it needs a model and a set of values for their limits, and you can’t make a guess until you have both of those things, a senior research scientist Greg Corrado at Google Research, told reporters.

“The numbers of limits a quantum computer can keep, and the number of operations it can hold, are very minute,”

That’s not to say that other companies aren’t planning about applying quantum computing to deep learning. Last month, two workers of protection contractor Lockheed Martin, which also bought a D-Wave quantum computer, published a paper documenting their hard work to use a D-Wave machine to assist train a deep neural network.