I Have been following D-Wave Quantum Computing for some time. And yes, if this plays out, it may well revolutionize aspects of machine learning. D-Wave has been used by NASA and Google. See my previous posts regarding D-Wave.
Quantum Leap: D-Wave's Next Quantum Computing Chip Offers a 1,000x Speed-Up
TechRepublic (09/16/16) Nick Heath
D-Wave's upgraded quantum processor, which is scheduled for release in 2017, promises to manage about 2,000 quantum bits (qubits) and solve certain problems 1,000 times faster than the existing D-Wave 2X processor. "We've been on a trajectory, which has been doubling the number of qubits pretty much every year," notes D-Wave's Colin Williams. He says the D-Wave system harnesses various atomic behaviors, including entanglement and state superposition, to solve a spectrum of complex computational challenges. The systems are specifically designed to conduct unconstrained binary optimization, along with related sampling problems.
D-Wave says the specialized work of the upgraded processor could be particularly applicable to training unsupervised machine-learning models. Williams thinks this aspect of the processor could have the greatest benefit, helping in the creation of "a machine that, having trained it, can...generate new data that is statistically indistinguishable from the kind of data on which it was trained." Ultimately, Williams envisions the D-Wave chip working alongside classical processors rather than supplanting them, and he rejects the notion that the chip's specialized nature limits its utility. "It's actually been shown that the one thing it does, can be used in lots of different fields," Williams says. .... "
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment