72 replies to this topic

[ThePhoeron]

Quantum computation is an interest of mine, and I've been following D-Wave in particular since 2011. Also, I registered with them when they launched their developer portal, so I have their Python pack/black box simulator, and play with it from time to time (when my day job allows me the time and mental energy to work on personal programming projects).

I would suppose that any potential applications for anti-aging research would ultimately be similar as with a 'classical' computer or supercomputer. Reduce the problem at large to a number of manageable problem sets. The advantage of quantum annealing and the D-Wave hardware is that it allows for efficient calculation of the types of problem sets that would challenge even our best supercomputers, in time and energy. So for problems dealing with a lot of probabilities and complex systems with numerous unknown variables, like anti-aging research, quantum energy programming is actually quite ideal.

Since this interests me, I will take some time to browse for likely problem sets in this area that could serve as good candidates. The real challenge is getting enough relevant data (not to mention the $10 million price tag on the D-Wave One, last time I checked).

[manofsan]

Well, is there any information/tutorials available online to show how to frame bioinformatics problems in a way that the D-Wave can process them?

D-Wave may be expensive, but so are mainframes and they were made accessible by timeshare long ago. I'd likewise imagine that D-Wave machines could be made accessible via cloud services. Actually, they should come up with a webservices framework to allow people to access their machine via the internet.

Is Python the only language they will accept computational job submissions in? Why not Java, etc?

[ThePhoeron]

Well, is there any information/tutorials available online to show how to frame bioinformatics problems in a way that the D-Wave can process them?

D-Wave may be expensive, but so are mainframes and they were made accessible by timeshare long ago. I'd likewise imagine that D-Wave machines could be made accessible via cloud services. Actually, they should come up with a webservices framework to allow people to access their machine via the internet.

Is Python the only language they will accept computational job submissions in? Why not Java, etc?

I think they designed their dev pack for Python to make quantum energy programming more accessible, but the Black Box itself was written in C++ and wrapped in an 'easy_install' package with a few extra utility functions. I was working on a Lisp cffi package, but since they won't release their source or header files, it quickly became an exercise in futility. Porting to Java would be a little impractical, though---java runs on the JVM, not on the bare metal, so a java-based Black Box would have two levels of virtualization and thus severely compromise the performance that you get out of the python pack. If you have the actual hardware though, you have access to the C++ API.

There are a few tutorials that you could apply to bioinformatics with a bit of stretching, I think, although the Molecule MCS is probably the closest. Most of the tutorials were designed with the AI community in mind---Hadamard matrix search, neural nets, weighted graphs, etc.

You can see their tutorials here: http://www.dwavesys....-tutorials.html But they are no longer accepting new developers for the python pack beta.

It's funny that you mention cloud services though, because that was a promised feature, the Quantum Cloud (they mention it under their Dev Portal). I haven't heard back about any progress with that service, though, in almost a year. They've apparently been too busy with some unspecified US Government projects.

[manofsan]

We really need to jump onto this bandwagon right away, now that it's turning out to be legit.

Wasn't it some British grad student who came up with the idea of using quantum physics to solve molecular bonding problems, which are themselves a result of quantum physics?

Anyway, I was thinking that this technology could be used for finding protein folding solutions, etc.

Sparse coding problems for better AI also seem to be a big new thing, which Google, Microsoft, etc are all busily pouring research money into.

I know it sounds like a frivolous thing to say, but my parallel computing prof loves to emphasize how all this new GPU-computing wouldn't have been possible without hardcore game enthusiasts pouring their money into driving that technology forward. So likewise, I'd wonder if quantum computing technology couldn't somehow be hooked up with the gaming market, to unleash the flood of money and developmental creativity.

I'd also wonder if this tech might not also be useful for stock trading purposes. Have any of these highly-paid "quants" looked into quantum computing technology?

At the very least, someone needs to at least come up with D-Wave version of FOLDING@HOME, as a tangible demonstration of how this technology can bring new benefits to the masses.

[manofsan]

Seems that Harvard researchers have already used D-wave for biological problems:

http://www.gizmag.co...computer/25558/

Meanwhile, D-wave is racking up an impressive ranking:

http://www.gizmag.co...-ranking/27476/

Edited by manofsan, 14 May 2013 - 04:24 AM.

[mikeinnaples]

The fledgling technology of quantum computing being displayed by d-wave is the spark that is going to ignite singularity. Even if you don't buy into the whole concept of singularity, the implications alone that this has for *our* cause here at Longecity is staggering.

[manofsan]

What seems apparent to me, is that the more qubits this thing grows to have, the more accurate or confident its results will be. Beyond a certain size threshold, this thing could work quite accurately and reliably.

If our powers of quantum manipulation and measurement continue to improve, I was wondering if this could one day lead to exploitation of Bell's Inequality, to allow instant FTL communication. You know, like the Ansible from Ender's Game.

[solarfingers]

http://www.hpcwire.com/hpc/508991.html

I wonder how long it will be till I have one running in my living room?

Forget the living room! I want one running in my head!

[ihatesnow]

http://www.scienceda...30620071525.htm

Edited by ihatesnow, 20 June 2013 - 01:25 PM.

[NeuroGuy]

The fledgling technology of quantum computing being displayed by d-wave is the spark that is going to ignite singularity. Even if you don't buy into the whole concept of singularity, the implications alone that this has for *our* cause here at Longecity is staggering.

You're spot on with both assertions. However, the staggering implication for our cause here at Longecity is that the technological singularity will most likely be the end of that cause; in other words, the end of human life. Oxford philosopher Nick Bostrom has written extensively on this, with an excellent resource being his recent book, Global Catastrophic Risks. There and elsewhere, Bostrom highlights the science community's conspicuous underestimation of the inherent risk of such an event, which coincides with the prevailing unreasonable faith in a materialistic utopia brought on by a genuine strong AI.

To date, I've yet to find a rational basis for asserting that such an entity would have any fundamental obligation to value human life, much less foster the development or even preservation of it. That leaves an attitude of indifference or hostility, either of which would ultimately result in the implacable demise of humanity. Moreover (and I mean no offense if this applies to you), any assertion that humanity could somehow control such a recursive intelligence is dangerously naive, on the scale of believing that a colony of ants could somehow control a growing human infant with no prior knowledge of the workings of an adult human mind.

[xks201]

Cool so lets live in fear and halt all scientific progress because computers will become our slave masters. Computers have no intuition of self preservation driven into their dna like we do. Please.

[A941]

However, all precautions to prevent any possible negative outcome should be considered and taken.

[stponky]

Pretty good talk. Makes me a little more hopeful about a general purpose quantum computer.