[ExI] D-Wave's Quantum Computer

[Anders Sandberg]

On 18/05/2013 16:35, Mike Dougherty wrote:
> On Sat, May 18, 2013 at 11:13 AM, Rafal Smigrodzki
> <rafal.smigrodzki at gmail.com> wrote:
>> How would an algorithm designer call the computation of network
>> structure from network traffic? Is it similar to QUBO?

Good question. I have a suspicion that there might be useful speedups, 
but it might be that the problem is similar to the ones that normal 
stochastic algorithms do just as well.

>>
>> If yes, then adiabatic computing might greatly reduce the amount of
>> data needed to upload a person.
> How many qubits will you need to model the neural network of a person?

That depends on what algorithm is used. Right now we do not have any 
quantum algorithm for running neural networks - there are some quantum 
neural network papers, but they are ANNs that do not correspond to real 
neural networks. It might simply be that Hodgin-Huxley compartment 
models lack quantum speedups if we are unlucky.

-- 
Anders Sandberg,
Future of Humanity Institute
Philosophy Faculty of Oxford University