I am studying how to apply neural networks to the problem of Quantum State Tomography and I got confused when it comes to decide if this is a supervised or unsupervised learning problem.

At first, I came across the usage of Restricted Boltzmann Machines in the domain of generative models, which are used for unsupervised learning, which seems right since the state tomography is a problem where I do not have access to the state in question and the only information available are measurement outcomes: thus, I am feeding my learning algorithm with unlabeled data, that is, in an unsupervised fashion.

But then I found this paper and also this one describing QST in the PAC-learning framework, which as far as I know is a framework applied to supervised problems only, but the paper seems ok since they were dealing with a training set of $(E, Tr(E\rho))$ where $E$ are measurement operators and $Tr(E\rho)$ are the trace of the measurement operator times the density matrix. The elements of the training set were drawn i.i.d from a probability distribution.

My problem here is: in QST the probability distribution and the density matrix are unknown. So how can I formulate QST as pac-learning and how can it be suited into the framework of a supervised learning algorithm? Or it depends on what kind of algorithm I am using?

RBMs are not the only approach. Nowadays, researchers are using deep neural networks and a lot of generative modeling, the most promissing being the usage of Conditional GANs.

What am I missing?

  • $\begingroup$ can you link the papers you're referring to? $\endgroup$
    – glS
    Mar 9 at 13:30
  • $\begingroup$ @glS yes! Just eddited the question with them $\endgroup$
    – Dimitri
    Mar 11 at 13:43

1 Answer 1


To me, QST (at least the nonadaptive kind) is more like a parameter estimation problem - a more constrained type of learning problem than ordinary supervised ML.$^1$

The typical setup for parameter estimation starts with an unknown variable $\theta$ sampled from a prior distribution $p(\theta)$. This $\theta$ is fixed for the duration of the learning task, and (for simplicity) induces a distribution of observations $$ p(\{(x_1, y_1), \dots, (x_n, y_n)\}|\theta):= p(B|\theta), \tag{1} $$ where $B$ represents the "training data", i.e. learner's observations. The learner's goal is, upon sampling observations from $p(B|\theta)$, to output an estimate $\hat{\theta} = \hat{\theta}(B)$ that is close to $\theta$. This might look like minimizing a distance $d(\theta, \hat{\theta})$ where the function $d$ does not depend on $p(B|\theta)$.

In basic QST, the observations look like $B=\{(E_i,\langle \hat{E_i}\rangle_{\rho} )\}_{i=1}^n$, where $E_i$ denotes a measurement setting and $\langle \hat{E_i}\rangle_{\rho}$ is an empirical estimate for the result of that measurement (subject to statistical noise, other experimental errors, etc). To connect this to parameter estimation, the variable $\theta:=\rho$ consists of all of the entries of $\rho$, and so $\hat{\theta}:= \hat{\rho}$ represents the learner's best guess for $\rho$. Typically we consider $d(\rho, \hat{\rho}) = \lVert \rho - \hat{\rho} \rVert_1$, and the learner wants to find $\hat{\rho}$ that minimizes this distance. If there's a prior distribution of states $p(\rho)$, the learner might try to minimize \begin{equation} \mathbb{E}_{p(\rho)} \left(\mathbb{E}_{p(B|\rho)} \lVert \rho - \hat{\rho}(\{(E_1,\langle \hat{E_1}\rangle_{\rho}), \dots, (E_n,\langle \hat{E_n}\rangle_{\rho})\}) \rVert_1\right) \tag{2} \end{equation}

The reason this differs from supervised learning is that the term in the parentheses above depends explicitly on $\rho$. In most supervised learning problems, you care less about learning the exact value of an underlying parameter, and more about predicting each new label $y_i$ given $x_i$.

Given the above, phrasing QST as a supervised learning problem is kind of sketchy, but I'll try anyways. Here's one possible setup: Suppose $\rho$ is still sampled according to $p(\rho)$, but that instead of learning $\rho$, you want to be able to predict measurement outcomes for randomly sampled measurements $E_i$. Mathematically, you want to come up with a function $f$ that minimizes \begin{equation} \mathbb{E}_{p(\rho)} \left(\mathbb{E}_{p(E)} |f(E) - \langle \hat{E}\rangle_{\rho}|\right) \tag{3} \end{equation} This now resembles a supervised learning problem. In fact, its almost identical to linear regression: You randomly sample pairs $(E_i, y_i)$ where $y_i:= \langle \hat{E}_i, \rho \rangle$, and your goal is to predict $y=\langle E, \rho \rangle$ with respect to the distribution of $E$.

This problem should not be much harder than QST, since if we had a good estimate $\hat{\rho}$ for $\rho$, then we could compute $\text{tr}(\hat{\rho}E_i) = y_i$. On the other hand, it might be that our distribution over measurements only contains a handful of measurements $E$ so that full knowledge of $\hat{\rho}$ is simply unnecessary for the task (this is some of the intuition behind classical shadows techniques).

$^1$ The way I use terminology here (learning, prediction, parameter estimation) isn't necessarily standard - its always better to just explain the learning setting in detail.

  • $\begingroup$ it was a great answer, very well argued. Thanks for the time $\endgroup$
    – Dimitri
    Mar 11 at 14:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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