9
$\begingroup$

There is obviously a great deal of work happening at the moment on variational quantum algorithms. However, I'm struggling to understand why exactly are they considered promising? Looking through some papers and review articles (such as this one https://arxiv.org/abs/2012.09265) I still do not get a clear idea.

As far as I understand there aren't much rigorous results on performance of these algorithms, similarly to many classic machine learning approaches. It is often argued that variational algorithms are great for NISQ as they require much less resources than a typical fault tolerant algorithm. However, without performance guarantees this looks a bit like searching under a lamppost to me. Why do we expect to actually find anything (=quantum advantage) there?

I also expect that some of the frequently mentioned applications of variational algorithms are more speculative than others. So, what are the most solid reasons to believe that variational algorithms can outperform classical? Are there any estimations for the amount of resources (number of qubits, gate fidelities etc) needed to achieve them?

$\endgroup$
3
  • 2
    $\begingroup$ There's a recent paper arxiv.org/abs/2106.12627 that shows this general "style" of computation, where a classical computer is assisted by something capable of sampling quantum distributions, can learn things more efficiently than just classical alone. Not sure if it addresses your question, since it's not specifically about VQE but it's sorta close. $\endgroup$ Jul 10 at 15:58
  • $\begingroup$ @Craig Gidney Oh, I saw that paper but the authors frame it a bit differently, as supervised vs unsupervised learning. That made me think that they are talking entirely about classical ML but the way you say it gives an interesting angle, thanks! $\endgroup$ Jul 10 at 16:47
  • $\begingroup$ related topic here and here $\endgroup$
    – narip
    Jul 14 at 4:31
5
+50
$\begingroup$

"As far as I understand there aren't many rigorous results on performance of these algorithms, similar to many classic machine learning approaches."

You are correct in that, unlike Grover's algorithm where we can prove that a search that would cost $\mathcal{O}(N)$ on a classical computer can be done with only $\mathcal{O}(\sqrt{N})$ on a quantum computer, "variational quantum algorithms" tend to be heuristics with far less for us to say in terms of theorems about their exact cost.

Now consider as an example, the "VQE" algorithm for estimating the energy of a quantum mechanical system (for example a molecule). This is essentially the Ritz method from at least 112 years ago: vary the parameters of the wavefunction in pursuit of lower and lower values of the function:

$$ \tag{1} \frac{\langle \psi | H| \psi \rangle }{\langle \psi|\psi \rangle}, $$

since Eq. 1 is provably an upper bound on the true energy of the system. This is already the way people have been estimating energies of quantum mechanical systems (for example if answering the question of whether glucose has more energy than fructose) on classical computers for many decades, and the VQE provably improves the efficiency of the overall procedure from a computational complexity perspective, because the cost of calculating Eq. 1 on a classical computer grows exponentially with the dimension of $H$, but on a quantum computer the cost would only grow polynomially.

Basically, VQE can exponentially speed up the most expensive part of the Ritz method from 112+ years ago, which for many problems is already the state-of-the-art method for classical computers.

Unfortunately the way VQE is usually described, the evaluation of Eq. 1 is done on a quantum computer (with provable exponential speed-up over classical computers), but then the energy and the parameters of the wavefunction are fed into a classical optimizer which tries to give improved parameters which will lead to a lower (and therefore better) estimation of the energy. Because the Hamiltonians are usually very sparse, the cost of calculating Eq. 1 on a classical computer is usually not actually exponentially scaling unless we have an exponentially scaling number of parameters in the wavefunction, which would mean that the cost of the "classical optimization" component of the VQE procedure would essentially make it worthless to do the whole procedure on a quantum computer (because an exponentially large number of parameters would be fed into the classical optimizer to get updated into improved parameters, but if a classical computer is capable of storing that many parameters, then the classical computer might as well also calculate Eq. 1, i.e. just do the whole VQE procedure classically like we've been doing since 112+ years ago!). Also, if you can calculate Eq. 1 with exponential speed-up on a quantum computer, you should be able to calculate the energy in a more direct way such as with quantum phase estimation or other methods in the Hamiltonian simulation category.

So then why did such a buzz around the term "VQE" arise since 2014?

Let's now talk about the coupled cluster method. This is a polynomially scaling classical algorithm for estimating quantum mechanical energies, and it is used extremely widely in quantum chemistry, where people use it to solve real-world problems about predicting the behavior of chemicals on a computer before doing dangerous experiments in a lab (unlike factoring numbers or cracking a Deutsch-Jozsa blackbox). It has widely been called the "gold standard" of quantum chemistry. Coupled cluster is not in general variational, and this is sometimes a concern for quantum chemists. Variational coupled cluster (vCC) and unitary coupled cluster (uCC) do exist as algorithms for classical computers, but are not considered practical, and the early VQE papers (for example in 2017) promoted the fact that quantum computers executing VQE could do uCC (meaning, an improved version of the "gold standard" of quantum chemistry). But unfortunately uCC this is just an improved version of a polynomially scaling energy estimation, whose advantage arises because there's only a polynomially scaling number of parameters in the wavefunction model, which is also a disadvantage because it doesn't have the exponentially scaling number of parameters required to get the "true" energy (in other words, it just gives an "estimation" of the energy which may or may not be slightly better than what people are doing on classical computers already).

While the above paragraph might be a bit depressing, the popularity of VQE with uCC circa 2017 led also to new methods like the hardware efficient ansatz in 2017 and qubit coupled cluster in 2018. The former created a huge buzz in the pop-sci media, not because VQE is promising, but because it was one of the world's biggest hardware companies (IBM) demonstrating an attempt at a solution of a real-world problem on real quantum hardware (a big deal back in 2017!) and publishing it in Nature; and the latter 2018 paper was actually very interesting science in itself.

$\endgroup$
4
  • $\begingroup$ Hi, thanks for the thorough reply and for the references, very helpful! Several follow-up questions. (i) Could you give a reference for the statement that computing eq.(1) is provably faster on a quantum computer? To my understanding this is related to the sampling from quantum distributions, which is subtle to prove (depends on the Hamiltonian, allowed errors etc.) $\endgroup$ Jul 19 at 10:09
  • $\begingroup$ (ii) Sorry, but I still do not see a clear-cut answer to my question! I'm still very uncertain. You seem to confirm that there is no complexity-theoretic backup for the current variational algorithms. What are the (best case) expectations then? That given sufficient amount of quantum resources and efficient classical optimization methods the variational algorithms will just work? Or let's put it pragmatically. I need to solve some real-world chemistry problem and ready to invest a lot in the research. What could make me invest in the quantum computing and not in classical algorithms? $\endgroup$ Jul 19 at 10:13
  • $\begingroup$ (i) Eq. 1 involves a vector with $2^n$ classical elements, which can be entirely represented by only $n$ qubits. (ii) If you are interested in real-world chemistry problems, I do not recommend for you to invest any money in quantum computing unless you have enough of it to be okay with the risk that you won't get a return for decades. $\endgroup$ Jul 19 at 19:13
  • $\begingroup$ (i) The "curse of dimensionality" arguments are actually not that strong, aren't they? There are broad classes of quantum systems/circuits that one can simulate efficiently on a classical computer. And to prove that there are classes that cant not would amount to proving $P\neq BQP$ (which is unsettled) would not it? (ii) OK, point taken, thanks. $\endgroup$ Jul 20 at 9:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.