New Book from OReilly: Programming Quantum Computers

[jedishrfu]

OReilly published a new book, Programming Quantum Computers.

From my brief scan of it, the book looks pretty comprehensive. They use circle notation for the Qubit states and have a simulator (QCengine) to go along with the examples on github. The programming language for their algorithm descriptions is JavaScript.

The book focus is on programmers interested in the technology but not so much the math and physics behind it.

https://www.amazon.com/dp/1492039683/?tag=pfamazon01-20

 

[Borg]

Javascript should be a perfect fit since you never know what the variables are anyway.

 

[Vanadium 50]

I am shocked that the cover isn't a cat. Or a half-faded cat.

 

[Borg]

Maybe it changes after you read it.

 

[anorlunda]

Oh, thanks for sharing @jedishrfu . I asked that question here on PF a long time ago, and never got an answer.

https://www.physicsforums.com/threads/making-a-quantum-computer-do-shors-algorithm.873585/

 

[Greg Bernhardt]

Holy moley, I didn't understand anything in that preview outside the intro pages lol. It's going to be a whole new world when quantum computing goes commercial.

 

[jedishrfu]

Maybe not. I think itll be like using a math coprocessor like in the early day of the micro.

 

[anorlunda]

Maybe not. I think itll be like using a math coprocessor like in the early day of the micro.

Blast from the past. That made me remember the Floating Point Systems add-on box we used in the late 70s. For the benefit of younger members, it was an external CPU that could do floating point calculations and array processing in the days when affordable CPUs did floating point with software.

Compare it to today's GPUs. Programming a GPU also sounds very alien.

 

[Baluncore]

I am shocked that the cover isn't a cat. Or a half-faded cat.

Extract; page 271.
The animal on the cover of Programming Quantum Computers is the musky octopus (Eledone moschata), a sea creature found at depths of up to ¼ mile in the Mediterranean Sea and the coastal waters of western Europe.
...
While the musky octopus’s current conservation status is designated as of Least Concern, many of the animals on O’Reilly covers are endangered; all of them are important to the world.
The cover illustration is by Karen Montgomery, based on a black and white engraving from Dover’s Animals.

 

[anorlunda]

Schrödinger's octopus

 

[jedishrfu]

The octopus makes sense to represent quantum mechanics because once it creates its inky cloud as a predator approaches then you can’t know where it is exactly anymore.

 

[heff001]

OK, Qbits driven machines that can solve problems that can take hundreds of years to solve and do it in ~2 weeks. Why isn't that the focus - that "is" the topic, not the cover.

 

[anorlunda]

OK, Qbits driven machines that can solve problems that can take hundreds of years to solve and do it in ~2 weeks. Why isn't that the focus - that "is" the topic, not the cover.

They have the potential to do that, but first they must be correctly programmed. They are also capable of arriving at incorrect answers in record time if incorrectly programmed.

Quantum computing is not the topic, programming them is.

 

[jedishrfu]

One thing folks forget here is that the answer doesn’t just pop out of the QC after one run. It’s a statistical process where the program is rerun many times maybe hundreds of times and the answer with the most number of matches wins.

We can check the answer classically for some problems rather quickly but finding it classically via an exhaustive search would take forever. The traveling salesman problem is a good example of one such problem.

 

[Vanadium 50]

By "check" you mean check that the solution is valid, not that the solution is optimal, right?

 

[jedishrfu]

Yes, as in the check is in the mail.

 

[anorlunda]

We can check the answer classically for some problems rather quickly

That's true, but the operative word is *some*. There will be other problems where we have no practical way to check the answers because checking by conventional computers won't work. If that is the case, then confidence in the programming may be the only evidence we have that the answers are correct.

We have not yet experienced quantum bugs.

That sounds like a ripe field for computer scientists. Correctness proofs for quantum programs.

 

[jedishrfu]

Also there are many problems suited for a classical computer but could never be programmed on a quantum computer at least from what I've seen so far.

This is a lot like the parallel computing paradox where there are some problems that can't be parallelized but we keep searching for ways to do it.