Quantum Computing

[lpetrich]

D-Wave Systems
IBM Quantum Computing

Quantum computing has been hyped for a long time, and the first usable quantum computers have gone on the market in the last few years.

Quantum computers use quantum bits or qubits, and use quantum-mechanical effects for fast computation. The D-wave machines implement qubits with superconducting loops. The amount of magnetic flux (magnetic field integrated over area) going through each loop is quantized, thus having amounts that are some multiple of a quantum of magnetic flux. Qubit loops that are near each other can interact, thus making quantum-mechanical entanglement and mixed states.

A big difficulty with both the D-wave and the IBM sites is that they are not very explicit about their model of computation. I did a lot of searching in their sites and it was hard to find anything. The D-wave machine apparently uses a "binary quadratic" model.

Minimize D(x,a,b) = (sum over i of a_i*x_i) + (sum over i,j of b_i,j*x_i*x_j)

Here, the x's are the qubit values, 0 or 1, and the a's and b's real numbers.

So if one can turn one's problem into a binary-quadratic problem, one has got it made.

For n qubits, the total number of possible values of D is 2ⁿ, and quantum effects make possible much faster searching than a classical-limit-computing full search. Such fast solution of exponential-time or factorial-time problems is a major reason for interest in quantum computing. Classical-limit computing has several relatively fast algorithms for approximate solution of such problems, but this speed has a price. These algorithms that will discover some local minimum with no guarantee that it will be a global minimum.

 

[excreationist]

Quantum Leaps in Quantum Computing?

New “qubit” designs could enable more robust quantum machines

www.scientificamerican.com

In principle, a 300-qubit quantum computer could perform more calculations at once than there are atoms in the observable universe.

(2^300 is greater than the 10^80 atoms in the observable universe)

26:40 .....last year we were the first company in the world to break the 100 qubit barrier for superconducting technology so we built a system with 117 with a processor called Eagle this year 433 next year 1127 after that we laid out our roadmap by 2025 we will have systems with at least 4000 qubits

I was looking into the travelling salesman problem:

Using Quantum Computing to Solve the Travelling Salesman Problem

Introduction Quantum computing is not just a buzzword and physical realisations of quantum computers have been produced over the last few decades, rightly drawing a lot of attention across a range of sectors. During our summer internship at PA Consulting’s Cambridge Innovation and Technology Centre,

www.linkedin.com

IBMs commercial quantum computer has 14 qubits and is only powerful enough to consider the travelling salesman problem for 4 cities. Conversely, D-Waves quantum annealer has 2048 qubits and it has the potential to find the shortest route for up to 9 cities

I don't think those 2048 qubits are fully connected with each other...

Each qubit in the annealer is not as powerful as each qubit in the gate-based quantum computer, but the massive difference is in how many we can use and how we use them which means the annealer is best suited to solve the TSP

 

[steve_bank]

Proprietary company information.

I don't know much about it. From what I have heard in popular reporting it is not sutable to genral computing as with a Turing Machine model.

I heard it said that quantum computing applied to encryption could make Internet encryption useless.

Computation requires a series of steps. Even at the quantum level there is a sequence of causation.


Is the integral of the magnetic filed a binary 1 or 0, or is it multilevel logic?

I can't kelp but think of the old magnetic core memory and wonder if there is any comparison.

There was also magnetic bubble memory.

Bubble memory - Wikipedia

en.wikipedia.org

Bubble memory is a type of non-volatile computer memory that uses a thin film of a magnetic material to hold small magnetized areas, known as bubbles or domains, each storing one bit of data. The material is arranged to form a series of parallel tracks that the bubbles can move along under the action of an external magnetic field. The bubbles are read by moving them to the edge of the material, where they can be read by a conventional magnetic pickup, and then rewritten on the far edge to keep the memory cycling through the material. In operation, bubble memories are similar to delay-line memory systems.