Looking at the paper it doesn’t appear to be outright bullshit to me. They use a standard approximation in quantum chemistry known as the Hartree-Fock approximation, which essentially allows you...
Looking at the paper it doesn’t appear to be outright bullshit to me. They use a standard approximation in quantum chemistry known as the Hartree-Fock approximation, which essentially allows you to more easily handle a large (i.e. many particle) system provided you’re ok with the needed approximations.
They then use a quantum algorithm to solve for the energy eigenstates of the resulting system. This is of course done on a quantum computer.
What may be bullshit is that it’s not clear to me that the system they were approximating is at all useful or interesting. Further, it doesn’t appear that this method offers an speed up or improvement over traditional methods that could be run on your laptop. So it is noteworthy for running on a QC, but not much else.
As for whether we’ll ever have useful quantum computers, I think it’s just entirely too early to tell. There was a lot of hype the last few years that made it seem like we were a couple of years away from fully scalable, fault tolerant computers, but I’m not sure that was ever realistic. The hype is dying down now and I think that gives the impression we’ve passed the peak and have nothing useful to show for it.
I personally think quantum computers will see useful applications in the next few decades, but being more specific than that is impossible.
I that most news is hype and that we are still far from anything useful, however this statement is simply not true. Ignoring much of the progress that has been made with understanding error...
We have made basically 0 progress on the intractable engineering problems associated with quantum computing.
I that most news is hype and that we are still far from anything useful, however this statement is simply not true. Ignoring much of the progress that has been made with understanding error correction in quantum algorithms, the field of quantum control has advanced great in the last ~20 years, which is essentially the life span of the field anyway. Just 20 years ago even controlling a quantum system like a Bose-Einstein condensate was hard enough that doing so won people a Nobel prize, and now it is "easy" enough that we can do it on the ISS. Quantum sensing and it's a huge field on its own. And not to mention the fact that we do have quantum computers today and didn't 20 years ago, granted the ones we have aren't excellent.
I bring these examples up because they tackle the same fundamental problem currently holding us up in quantum computing; namely, it is very very hard to precisely control quantum systems without decoherence, and for a long enough time to be useful. Difficult, but not impossible. I haven't seen anything to suggest so far that the fundamental problems are inherently intractable, and with enough funding I would expect to see modest incremental improvements that lead us into the realm of usability (in a NISQ sense, where there are some limited applications, but still not the universal supremacy promised by some advocates a few years back). And we're seeing the needed funding get acquired, too: the US alone is investing $625 million over 5 years in 5 different new quantum research centers. That's an insane amount. And these centers will be focusing not only on quantum computing, but things like quantum sensing as well, so I'm hopeful there will be enough cross germination to see some big breakthroughs.
Looking at the paper it doesn’t appear to be outright bullshit to me. They use a standard approximation in quantum chemistry known as the Hartree-Fock approximation, which essentially allows you to more easily handle a large (i.e. many particle) system provided you’re ok with the needed approximations.
They then use a quantum algorithm to solve for the energy eigenstates of the resulting system. This is of course done on a quantum computer.
What may be bullshit is that it’s not clear to me that the system they were approximating is at all useful or interesting. Further, it doesn’t appear that this method offers an speed up or improvement over traditional methods that could be run on your laptop. So it is noteworthy for running on a QC, but not much else.
As for whether we’ll ever have useful quantum computers, I think it’s just entirely too early to tell. There was a lot of hype the last few years that made it seem like we were a couple of years away from fully scalable, fault tolerant computers, but I’m not sure that was ever realistic. The hype is dying down now and I think that gives the impression we’ve passed the peak and have nothing useful to show for it.
I personally think quantum computers will see useful applications in the next few decades, but being more specific than that is impossible.
I that most news is hype and that we are still far from anything useful, however this statement is simply not true. Ignoring much of the progress that has been made with understanding error correction in quantum algorithms, the field of quantum control has advanced great in the last ~20 years, which is essentially the life span of the field anyway. Just 20 years ago even controlling a quantum system like a Bose-Einstein condensate was hard enough that doing so won people a Nobel prize, and now it is "easy" enough that we can do it on the ISS. Quantum sensing and it's a huge field on its own. And not to mention the fact that we do have quantum computers today and didn't 20 years ago, granted the ones we have aren't excellent.
I bring these examples up because they tackle the same fundamental problem currently holding us up in quantum computing; namely, it is very very hard to precisely control quantum systems without decoherence, and for a long enough time to be useful. Difficult, but not impossible. I haven't seen anything to suggest so far that the fundamental problems are inherently intractable, and with enough funding I would expect to see modest incremental improvements that lead us into the realm of usability (in a NISQ sense, where there are some limited applications, but still not the universal supremacy promised by some advocates a few years back). And we're seeing the needed funding get acquired, too: the US alone is investing $625 million over 5 years in 5 different new quantum research centers. That's an insane amount. And these centers will be focusing not only on quantum computing, but things like quantum sensing as well, so I'm hopeful there will be enough cross germination to see some big breakthroughs.
That is amazing! Is there any information on this account? Like how is it calling BS? AI? Experts in the field?