Abstract
Markov logic networks (MLNs) reconcile two opposing schools in machine learning and artificial intelligence: causal networks, which account for uncertainty extremely well, and firstorder logic, which allows for formal deduction. An MLN is essentially a firstorder logic template to generate Markov networks. Inference in MLNs is probabilistic and it is often performed by approximate methods such as Markov chain Monte Carlo (MCMC) Gibbs sampling. An MLN has many regular, symmetric structures that can be exploited at both firstorder level and in the generated Markov network. We analyze the graph structures that are produced by various lifting methods and investigate the extent to which quantum protocols can be used to speed up Gibbs sampling with state preparation and measurement schemes. We review different such approaches, discuss their advantages, theoretical limitations, and their appeal to implementations. We find that a straightforward application of a recent result yields exponential speedup compared to classical heuristics in approximate probabilistic inference, thereby demonstrating another example where advanced quantum resources can potentially prove useful in machine learning.
Introduction
Graphical models combine uncertainty and logical structure in an intuitive representation. Examples include Bayesian networks, Markov networks, conditional random fields, and hidden Markov models, but also Ising models and Kalman filters. Their main advantage is the compactness of representation, stemming from capturing the sparsity structure of the model and independence conditions among the variables reflected in the correlations. The graph structure encompasses the qualitative properties of the distribution. Exact probabilistic inference in a general Bayesian or Markov network is #Pcomplete^{1}, which is why one often resorts to Markov chain Monte Carlo (MCMC) Gibbs sampling to approximate exact probabilistic inference. However, the task remains computationally intensive even with MCMC.
Graphical models belong to a school of machine learning that emphasizes the importance of probability theory. Firstorder logic on the contrary comes from the symbolist tradition of artificial intelligence and it relies on inverse deduction to perform inference. Markov logic networks reconcile the two schools, and in one limit, they recover firstorder logic^{2}. A Markov logic network is essentially a template for generating Markov networks based on a knowledge base of firstorder logic. MCMC Gibbs sampling can be used in the same way as in ordinary Markov networks to perform approximate probabilistic inference, but it suffers from the enormous number of nodes that are generated by the template.
There has been a recent surge of interest in using quantum resources to improve the computational complexity of various tasks in machine learning^{3,4,5,6}, similar to what one aims to achieve more generally in the fields of quantum communication^{7} and quantum computation^{8,9,10,11}. This approach has been successful in training Boltzmann machines, which are simple generative neural networks of a bipartite structure—a set of hidden and a set of visible nodes—where the connectivity is full between the two layers. Edges carry weights and these are adjusted during training. We can view Boltzmann machines as Markov networks with a special topology, in which the largest clique has size two. One method employed for training Boltzmann machines^{12,13,14} is quantum annealing. It is a global optimization method that relies on actual physical phenomena and it can be used to generate a Gibbs distribution. For all current quantum annealing approaches to Gibbs sampling, restrictions on the topology of the physical hardware remain the main obstacle, which is why the limited clique size of the Boltzmann machines is attractive. An alternative approach of training Boltzmann machines is by using Gibbs state preparation and sampling protocols, which can also exploit the structure of the graph and achieve polynomial improvements in computational complexity relative to its classical analogue^{15}.
Here, we go beyond the training of Boltzmann machines and consider more general Markov logic networks, keeping the expressiveness of firstorder logic and concentrate on inference, rather than training. We analyze the usefulness of quantum Gibbs sampling methods to outperform MCMC methods. The runtime of quantum Gibbs sampling algorithms is sensitive to both the connectivity structure and the overall number of subsystems. Methods of lifted inference can be used to address these issues.
Probabilistic Inference and Lifting
Markov networks are undirected graphical models that offer a simple perspective on the independence structure of a joint probability distribution of random variables, and the task of probabilistic inference based on this structure^{1}. Nodes of the network are random variables and edges between nodes imply influence or direct correlation, that is, lack of conditional independence. Instead of conditional probabilities on parent nodes, as in Bayesian networks, Markov networks operate with unnormalized factors f_{j}, that is, functions that map from subsets of the random variables to nonnegative reals. The factors are defined over the cliques of the graph. To obtain a valid joint probability distribution over the random variables from the factors, a partition function normalizes the unnormalized measure, so that the probability distribution takes the form , where x_{j} are subsets of x corresponding to the cliques and Z is the partition function. If P is a positive distribution over the random variables , we can associate a Gibbs distribution to the Markov network as , where the features g_{j} are functions of a subset of the state, and w_{j} are real weights.
In firstorder logic, constants are objects over some domain (e.g., Alice, Bob, … in the domain of people), and variables range over the set of constants in the domain. A predicate is a symbol that represents an attribute of an object (e.g, Smokes), or a relation among objects (e.g., Friends). An atom is a predicate applied to a tuple of variables or constants. A ground atom only has constants as arguments. These definitions apply to a function free language with finite size domains—technically, this is a strict subset of firstorder logic. A formula is constructed of atoms, logical connectives, and quantifiers over variables. A knowledge base is a set of formulas connected by conjunction. A world is an assignment of a truth values to each possible grounding of all atoms in a knowledge base. An essential task in a firstorder knowledge base is to check whether a formula is satisfiable, that is, there exists at least one world in which it is true.
To relax the rigid trueorfalse nature of firstorder logic, Markov logic networks (MLNs) introduce a real weight w_{j} for each formula f_{j} in a knowledge base^{2}. A Markov logic network is a set of pairs , representing a probability distribution over worlds as
where is the number of groundings of f_{j} that are True in the world ω. An MLN can be thought of as a graph over the set of all possible groundings of the atoms appearing in the knowledge base. The size of this graph is , where D is the maximum domain size, and c is the highest number of atoms in any of the formulas in the knowledge base^{16}. Groundings are viewed as connected if they can jointly appear in a grounding of some formula of the knowledge base. The ground network thus contains cliques, i.e., fully connected subgraphs, consisting of grounded atoms that jointly appear in the grounding of some formula. The maximum clique size k is given by the maximum number of atoms per formula. Table 1 summarizes how the structure of the firstorder knowledge base influences the characteristics of the generated Markov network.
MLNs belong to the class of methods known as statistical relational learning, which combine relational structures and uncertainty^{17}. An MLN essentially uses a firstorder logic knowledge base as a template to generate a Markov network by grounding out all formulas. An MLN can always be converted to a normal MLN, which has the following two properties: (i) there are no constants in any formula; (ii) given two distinct atoms with the same predicate symbol with two variables x and y in the same argument, then the domain of the two variables is identical. In the rest of this work we assume all MLNs to be given in this normal form. We further assume that skolemnization is applied to convert existential quantifiers to universal quantifier, which can be done in polynomial time in the size of a formula with no unquantified variables^{18}.
A main task in graphical models and in MLNs is probabilistic inference. One aspect of it is computing the partition function. The other aspect deals with the problem of assigning probabilities to or finding (the most) likely assignment of variables given evidence, that is, given a fixed assignment for a subset of its variables. This is a hard problem in general: the worstcase complexity of exact probabilistic inference of a graphical model is #complete and that of approximate inference is #hard^{1}.
For some common graphical models with a special topology, efficient exact probabilistic inference methods are known. Examples include belief propagation^{19} and the junction tree algorithm^{20}. In other cases, MCMC Gibbs sampling is often used for approximate inference to escape the worstcase complexity of exact inference. MCMC is hereby used to approximately sample from the distribution given in (1) or from a suitable conditional probability distribution conditioned on the evidence .
Graphical models often have symmetries that reduce the overall complexity of both exact and approximate inference. For instance, counting belief propagation exploits symmetries for exact inference^{16}, and orbital Markov chains do the same for approximate inference^{21}. Some of these methods have special extensions for MLNs, for instance, one can detect a subset of components in the ground network that would behave identically during belief propagation^{22}. It is worth exploiting the symmetries that emerge from firstorder logic and they are best exploited before grounding out, that is, symmetries should be addressed at the propositional level.
Approximate and exact probabilistic inference for firstorder probabilistic languages predates MLNs^{23,24,25}. The core idea is a form of coarse graining by grouping similar variables together. This idea was exploited in lifted firstorder probabilistic inference for MLNs^{26}. For hierarchically typed MLNs, one can move from coarsegraining over the highest level in a type hierarchy to more refined types^{27}.
Exploiting symmetries in the presence of evidence must be done with great care. Given evidence, the symmetries can become skewed, as random variables do not appear symmetrically in the formulas of the knowledge base^{28}. In this case, importance sampling helps^{29,30}, which clusters similar network components together given the evidence^{31}, and approximates the correct probabilities by an easier probability distribution and an estimated importance or weight of the error.
For most practical applications, either belief propagation or MCMC, augmented with some of the described techniques as appropriate for the problem at hand, is the method of choice for approximate probabilistic inference with MLNs. While often yielding useful results with an effort far smaller than the worst case complexity, they remain very expensive computationally and so more efficient alternatives are desirable.
Quantum Gibbs Sampling
The distribution (1) we would like to sample from can be thought of as the Gibbs distribution of a suitably constructed physical system. According to the rules of statistical mechanics, the probability to find a system in a certain state of configuration when it is in thermal equilibrium follows a Gibbs distribution. The distribution can thus be sampled by preparing a suitable physical system in a thermal equilibrium Gibbs state and then measuring its configuration. This is generally rather easy to do at high temperatures, but cooling to low temperatures typically becomes increasingly difficult. Thereby methods of quantum information processing can offer advantages over classical strategies. They open up fundamentally new ways to of preparing systems approximately in Gibbs states in a wellcontrolled way.
Going from the abstract definition of the probability distribution in (1) to a physical model can be done in the following way: We can think of as the “energy” of a system of n spin 1/2 “particles” in a quantum state . The states are then product state vectors in the Hilbert space with span the complex linear span. We can think of as the inverse of the “temperature” T of the system, times the Boltzmann constant k_{B} (other decompositions of the features are also possible). We can try to find a Hamiltonian H such that we can rewrite the probability distribution from (1) as follows
Thereby is the Hermitian conjugate of the state vector and is the partition function, where exp is the matrix exponential and tr the matrix trace.
In the concrete case of an MLN, the number of particles is equal to the number of all possible groundings of the atoms in the knowledge base underlying the MLN. The Hamiltonian H inherits the locality structure of the MLN: it can be written as a sum of local terms h_{l}, one for each clique of the MLN. More precisely, for each j the expression translates to a sum over local terms each acting on one of the cliques produced by grounding out f_{j} and acting on this clique like times the projector on the subspace of assignments to the atoms in the clique for which f_{j} evaluates to True. The local terms h_{l} of the Hamiltonian can be constructed from the truth tables of the the f_{j} and the sum over l in the decomposition of H collects all such terms for the different values of j. Figure 1 illustrates the matching concepts in MLNs and this description.
The number k of subsystems on which each such term acts nontrivially is bounded by the maximum number of atoms per formula and its operator norm is bounded by one . Hence (1) is the thermal Gibbs distribution of a system of n spin 1/2 particles with a socalled klocal Hamiltonian H. To prepare the system in a state that is suitable to sample from (1) it is sufficient to reach a high effective temperature if all weights are of moderate magnitude (no assignments are strongly suppressed), but it is necessary to cool to a low temperature if weights have a high magnitude (at least one assignment is strongly suppressed).
Computational Complexity
Quantum Gibbs sampling methods can be used to obtain samples from the Gibbs distributions of the type of systems described in the previous section. Typically these methods consist of two phases:
A preparation phase in which a quantum system is prepared in (a state close to) a state encoding information about the Gibbs state or such a state itself; and
A measurement phase, in which, by performing measurements on this state, samples from the Gibbs distribution are obtained.
The measurement phase is trivial, consisting only of local measurements and has complexity . The known quantum methods for Gibbs sampling differ in the kind of resources they require during the preparation, their expected improvement in runtime over classical methods, and the extent to and effort with which their performance for a concrete Gibbs distribution can be predicted.
The state prepared in the preparation phase is usually either close to a thermal Gibbs state^{32,33,34,35} at inverse temperature β of a given Hamiltonian H, or to a socalled pure thermal state^{36,37}, i.e., a pure state whose overlap with any energy eigenstates of H with energy E is proportional to the square root of the Gibbs weight exp(−βE). Recently, an algorithm for the computation based approximate preparation of thermal states of arbitrary klocal Hamiltonians has been proposed in ref. 35. For this algorithm a particularly favorable upper bound on the gate complexity—the scaling of the number of elementary operations in the preparation step—is known. This bound can be expressed in terms of the inverse temperature β, the local dimension d (d = 2 for Gibbs states corresponding to MLNs), the number of local terms in H, the gate complexity of time evolution under these terms (or the size of their support) and their strength, as well as the value of the partition function Z = tr(exp(−βH)), and the final distance to the thermal state ε.
Proposition 1 Assuming that the maximum size of the support of the local terms of the Hamiltonian H is constant and that for some constant α the number of terms in H is in , the overall complexity of the Gibbs sampling method from ref. 35 is in .
More generally it is enough if the gate complexity of time evolution under the local terms of H scales at most linear with n. When applying this to the graph structure generated by an MLN, α can be taken to be the maximum number of atoms in any formula, the maximum size of the supports of the local terms of H is equal to the maximum clique size, and the number of terms is the number of cliques in the MLN. So long as the maximum number of atoms in any formula is constant, the above scaling of complexity is achieved. It is important to note that the complexity does not directly depend on the maximal degree of the MLN.
This result improves upon the previously known methods in several respects, but in particular it improves the scaling of the runtime with 1/ε and β. In the natural parameters, the problem size n and the precision ε, this method yields an exponential improvement over the runtime of classical simulated annealing, which scales like , where δ is the gap of the Markov process, which in interesting cases typically is in ^{36}. However, the exponential dependence on n remains. The possibility of a logarithmic scaling with 1/ε was anticipated in refs 38, 39, 40. This scaling is particularly relevant when small probabilities are to be estimated with small relative error.
Following early works^{41}, several previous methods for quantum Gibbs sampling with improved scaling of complexity had been proposed^{32,33,36,37,38,39,40,42,43}. This in particular concerns the dependence of the runtime on the dimension of the Hilbert space or the inverse gap of a Markov chain , which was reduced from linear to square root by using techniques such as Szegedy’s quantum walks, the Grover/Long algorithm^{44,45}, phase estimation^{46}, or amplitude amplification^{47}. Algorithms that speed up the convergence of Markov Chains with quantum techniques^{36,37,38,39,40,42,43} often offer more flexibility than such more specific to the problem of preparing thermal states^{32,33,35}. In cases in which the gap a Markov chain is much larger than , they combine their quantum speedup with the advantage inherent in MCMC. However, the interesting cases are usually those in which and then both types of algorithms perform essentially equally well. A different method, based on the preparation of microcanonical states, was developed in ref. 34, but has an at least exponential scaling in βH.
If the Hamiltonian H has more structure and/or the effective temperature is high, more efficient special purpose procedures are available^{15,48,49}, which however are of limited relevance for inference in MLNs. In addition to that, there exists a quantum generalization of the Metropolis sampling algorithm^{50}, that however does not aim at achieving a speedup, but rather works around the sign problem in fermionic systems and makes MCMC techniques available for general local quantum Hamiltonians with noncommuting terms.
Can We Hope for Something Better?
As we have seen, quantum methods reduce the complexity of approximate Gibbs sampling quite drastically. Still, an exponential scaling with the number of all possible groundings of all atoms n remains, and the complexity diverges in the low temperature limit as β goes to infinity. A valid question is: Can we hope that future advances will remedy this? After all, the quantum Gibbs sampling methods presented above are able to do Gibbs sampling from Hamiltonians much more general than those that can arise from MLNs, like ones that have noncommuting terms, for example. Yet, the answer is probably negative. It is highly unlikely that any general purpose quantum algorithm for inference in MLNs exists that is efficient in cases with high weights (i.e., at low temperatures), as this would imply the existence of an efficient algorithm for solving satisfiability problems more general than 3SAT, which is known to be NPcomplete by the Cook–Levin theorem^{51}. Further, the log(1/ε) scaling of complexity is known to be optimal for Hamiltonian simulation^{52} and hence for any Gibbs sampling method based on it (the algorithm of ref. 52, like any other algorithm whose operations are written as linear combinations of unitaries can be thought of^{53} as a duality quantum computing algorithm^{54,55,56}; this is always possible as the unitary Pauli Matrices form a complete basis). The situation is different in the high temperature regime, where more efficient Gibbs sampling methods exist^{48}.
Computing the Partition Function with FirstOrder Lifting
The great advantage of working with lifting at the firstorder level is that we save potentially exponentially many groundings given the compact representation when we count the models in Eq. (1). There are trivial cases: for instance, when there are no shared variables between the atoms, then there is a closed form to calculate the number of satisfied groundings^{57}. Here we follow the outlines of lifted importance sampling^{29,58}, but without reference to an importance or proposal distribution: our aim is to reduce the complexity of the generated Markov network and potentially split it into disconnected graphs when computing the partition function. We run quantum Gibbs sampling on the smaller network and postprocess the result with some bookkeeping values to return the value of the partition function. Algorithm 1 summarizes the steps. Since the sampling is not based on a proposal distribution, the actual variance will depend on the error term that estimates the accuracy of the quantum Gibbs sampler. We follow the simplification steps from lifted importance sampling to cater to the critical parts of quantum thermal state preparation, but in principle, the sampling part of the algorithm can also use classical MCMC Gibbs sampling. For this reason, Algorithm 1 does not specify what kind of Gibbs sampling protocol we use.
If we have a normalform network as the input, that is, all domains have size one, we can run the Gibbs sampler and return the value of the partition function.
The first interesting case is if we detect a decomposer—this can be done in linear time—that is, a set of logical variables x such that (i) every atom in contains exactly one variable from x, and (ii) for every predicate R there exists a position such that variables from x only appear at that position. If we have a decomposer, can be simplified to that is obtained by substituting all variables in x by the same constant X in , , then converting the result to normal form. The partition function is calculated as .
The next structural simplification comes from isolated variables—one such variable in a predicate R at position m is exclusive to R in all formulas containing R. Let x denote all isolated variables of R and y the rest of the variables, and . We obtain a simplified MLN by generating the groundings of for , deleting the formulas that evaluate True or False, deleting all groundings of R, and normalizing the result. We get a combinatorial multiplier term to adjust the value of the partition function.
The final simplification is known as the generalized binomial rule, which relies on singleton atoms that do not appear more than once in the same formula. Given such an atom , we can simplify the MLN as , where is a truth assignment to all groundings of such that exactly groundings are set to . The simplified network is obtained by grounding all and setting all its groundings to match the assignment given by , deleting the formulas that evaluate or , deleting all groundings of , and normalizing the result. We can compute the partition function by , where is the exponentiated sum of the formulas that evaluate to , and is the number of ground atoms that are removed when removing the formulas.
If we cannot find any heuristics, we have to resort to fully grounding out an atom, normalizing the result, and continuing with the remaining expressions.
Probabilistic Inference Given Evidence
If we look at probabilistic inference given evidence, at the level of the quantum protocol, this can be done in at least two ways: First, to the Hamiltonian H one can add some strong local “clamping” terms, effectively forcing some of the assignments to the desired values. This is convenient from an implementation point of view, as it only requires few local changes in the Hamiltonian simulation procedure^{52} underlying the algorithm of ref. 35. However, it can be difficult to quantify the additional error due to the finite clamping strength and adding very strong clamping terms unfavorably affect the runtime of the algorithm. Second, one can construct the local terms h_{l} not from the full truth tables of the f_{j}, but instead use reduced truth tables given the evidence, to construct local terms h_{l} that act nontrivially only on the grounded atoms for which no evidence exists. This can only decrease the maximal weight (i.e., increase the temperature 1/β), decrease the number of terms (in case some of them become completely trivial), and reduce the number of sites n. Gibbs sampling with the new Hamiltonian is hence always at most as computationally costly as with the original Hamiltonian.
We can also use classical heuristics before employing the quantum protocol, as in the algorithm described in the previous section. For firstorder lifting methods, the presence of evidence is a problem, as it skews symmetries and potentially leads to a complete grounding out. To avoid this, ref. 31 proposed a distance function on the partially clamped network, and suggested a clustering to find clusters of similar groundings. All groundings in a cluster are replaced by their cluster center, reducing the overall network size to to , where r is maximum cluster size, compared to the original . This in turn reduces n in the overall complexity of the quantum Gibbs sampling protocol, as stated in Proposition 1.
Conclusions and Future Work
We hope that by fostering knowledge exchange between communities, for example concerning the typical properties of Gibbs distributions relevant for machine learning, progress towards more realistic and useful quantum algorithms can be made. In summary, we addressed the following aspects of probabilistic inference in MLNs:
We analyzed the computational complexity of the stateoftheart quantum Gibbs sampling protocol given the structural properties of MLNs and we argued the theoretical limits of the approach. A term in the computational complexity reduces exponentially, albeit the overall complexity remains exponential in the number of nodes.
Understanding the impact of the properties of the graph generated by an MLN on the computational complexity of quantum Gibbs sampling, we adapted a classical firstorder lifting algorithm to reduce the complexity of the network. The algorithm mirrors lifted importance sampling, but instead of using a proposal distribution, it uses either classical MCMC or quantum Gibbs sampling.
We studied the effects of evidence on quantum Gibbs sampling.
The protocols we considered rely on a universal quantum computer, which, given the hurdles in implementation, is still mainly of academic interest. We can, however, turn to methods that use current or near future quantum annealing devices, for instance, technology using quantum annealing with manufactured spins^{59,60}. In this technology, the distribution of excited states after annealing follows approximately a Boltzmann distribution^{12}, albeit one has to pay attention to estimating persistent biases and the effective temperature estimation^{13,14}. This technology was used, for instance, for learning the structure of a Bayesian network^{61}, but the restricted connectivity between the spins causes difficulties for arbitrary graph structures, in contrast to the methods discussed here. Recent progress allows embedding arbitrary graphs, albeit at a quadratic cost in the number of spins in the worstcase scenario^{62,63}, and there is also a proposal for a quantum annealing architecture with alltoall connectivity^{64}. Given the techniques described in this paper, it would be interesting to see whether we can achieve a scalable implementation with contemporary quantum annealing technologies, since machine learning demonstrations with this paradigm mainly focused on Boltzmann machines so far: MLNs have different topological features than Boltzmann machines, but they also have regularities that might allow an efficient embedding and subsequent inference.
Additional Information
How to cite this article: Wittek, P. and Gogolin, C. Quantum Enhanced Inference in Markov Logic Networks. Sci. Rep. 7, 45672; doi: 10.1038/srep45672 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
 1.
Koller, D., Friedman, N., Getoor, L. & Taskar, B. Graphical models in a nutshell. In Getoor, L. & Taskar, B. (eds.) Introduction to Statistical Relational Learning (MIT Press, 2007).
 2.
Richardson, M. & Domingos, P. Markov logic networks. Machine Learning 62, 107–136, doi: 10.1007/s1099400658331 (2006).
 3.
Schuld, M., Sinayskiy, I. & Petruccione, F. An introduction to quantum machine learning. Contemporary Physics 56, 1–14, doi: 10.1080/00107514.2014.964942 (2014).
 4.
Wittek, P. Quantum Machine Learning: What Quantum Computing Means to Data Mining (Academic Press, New York, NY, USA, 2014).
 5.
Adcock, J. et al. Advances in quantum machine learning. arXiv:1512.02900 (2015).
 6.
Biamonte, J. et al. Quantum machine learning. arXiv:1611.09347 (2016).
 7.
Pfaff, W. et al. Unconditional quantum teleportation between distant solidstate quantum bits. Science 345, 532–535, doi: 10.1126/science.1253512 (2014).
 8.
Tiecke, T. G. et al. Nanophotonic quantum phase switch with a single atom. Nature 508, 241–244, doi: 10.1038/nature13188 (2014).
 9.
Reiserer, A., Kalb, N., Rempe, G. & Ritter, S. A quantum gate between a flying optical photon and a single trapped atom. Nature 508, 237–240, doi: 10.1038/nature13177 (2014).
 10.
Ren, B.C., Wang, G.Y. & Deng, F.G. Universal hyperparallel hybrid photonic quantum gates with dipoleinduced transparency in the weakcoupling regime. Physical Review A 91, 032328, doi: 10.1103/PhysRevA.91.032328 (2015).
 11.
Wei, H.R., Deng, F.G. & Long, G. L. Hyperparallel Toffoli gate on threephoton system with two degrees of freedom assisted by singlesided optical microcavities. Optics Express 24, 18619, doi: 10.1364/OE.24.018619 (2016).
 12.
Adachi, S. H. & Henderson, M. P. Application of quantum annealing to training of deep neural networks. arXiv:1510.06356 (2015).
 13.
Benedetti, M., RealpeGómez, J., Biswas, R. & PerdomoOrtiz, A. Estimation of effective temperatures in a quantum annealer and its impact in sampling applications: A case study towards deep learning applications. Physical Review A 94, 022308, doi: 10.1103/physreva.94.022308 (2015).
 14.
PerdomoOrtiz, A., O’Gorman, B., Fluegemann, J., Biswas, R. & Smelyanskiy, V. N. Determination and correction of persistent biases in quantum annealers. arXiv:1503.05679 (2015).
 15.
Wiebe, N., Kapoor, A. & Svore, K. M. Quantum deep learning. arXiv:1412.3489 (2014).
 16.
Kersting, K., Ahmadi, B. & Natarajan, S. Counting belief propagation. In Proceedings of UAI09, 25th Conference on Uncertainty in Artificial Intelligence, 277–284 (2009).
 17.
Getoor, L. & Taskar, B. (eds.) Introduction to Statistical Relational Learning (MIT Press, 2007).
 18.
Van den Broeck, G., Meert, W. & Darwiche, A. Skolemization for weighted firstorder model counting. In Proceedings of KR14, 14th International Conference on Principles of Knowledge Representation and Reasoning, 1–10 (2014).
 19.
Pearl, J. Reverend Bayes on inference engines: A distributed hierarchical approach. In Proceedings of AAAI82, 2nd National Conference on Artificial Intelligence, 133–136 (1982).
 20.
Lauritzen, S. L. & Spiegelhalter, D. J. Local computations with probabilities on graphical structures and their application to expert systems. Journal of the Royal Statistical Society. Series B (Methodological) 50, 157–224 (1988).
 21.
Niepert, M. Markov chains on orbits of permutation groups. In Proceedings of UAI12, 28th Conference on Uncertainty in Artificial Intelligence, 624–633 (2012).
 22.
Singla, P. & Domingos, P. M. Lifted firstorder belief propagation. In Proceedings of AAAI08, 23rd Conference on Artificial Intelligence, vol. 8, 1094–1099 (2008).
 23.
Pasula, H. & Russell, S. Approximate inference for firstorder probabilistic languages. In Proceedings of ICJAI01, 17th International Joint Conference on Artificial Intelligence, 741–748 (2001).
 24.
Poole, D. Firstorder probabilistic inference. In Proceedings of IJCAI03, 18th International Joint Conference on Artificial Intelligence, 985–991 (2003).
 25.
De Salvo Braz, R., Amir, E. & Roth, D. Lifted firstorder probabilistic inference. In Proceedings of IJCAI05, 19th International Joint Conference on Artificial Intelligence, 1319–1325 (2005).
 26.
Kersting, K. Lifted probabilistic inference. In Proceedings of ECAI12, 20th European Conference on Artificial Intelligence, 33–38 (2012).
 27.
Kiddon, C. & Domingos, P. Coarsetofine inference and learning for firstorder probabilistic models. In Proceedings of AAAI11, 25th AAAI Conference on Artificial Intelligence, 1049–1056 (2011).
 28.
Ahmadi, B., Kersting, K., Mladenov, M. & Natarajan, S. Exploiting symmetries for scaling loopy belief propagation and relational training. Machine Learning 92, 91–132, doi: 10.1007/s1099401353850 (2013).
 29.
Gogate, V., Jha, A. & Venugopal, D. Advances in lifted importance sampling. In Proceedings of AAAI12, 26th AAAI Conference on Artificial Intelligence, 1910–1916 (2012).
 30.
Venugopal, D. & Gogate, V. Scalingup importance sampling for Markov logic networks. In Advances in Neural Information Processing Systems, vol. 27, 2978–2986 (2014).
 31.
Venugopal, D. & Gogate, V. Evidencebased clustering for scalable inference in Markov logic. In Machine Learning and Knowledge Discovery in Databases, 258–273 (2014).
 32.
Poulin, D. & Wocjan, P. Sampling from the thermal quantum Gibbs state and evaluating partition functions with a quantum computer. Physical Review Letters 103, 220502, doi: 10.1103/PhysRevLett.103.220502 (2009).
 33.
Chiang, C.F. & Wocjan, P. Quantum algorithm for preparing thermal Gibbs states  detailed analysis. In Quantum Cryptography and Computing, vol. 26 of NATO Science for Peace and Security Series  D: Information and Communication Security, 138–147 (2010).
 34.
Riera, A., Gogolin, C. & Eisert, J. Thermalization in nature and on a quantum computer. Physical Review Letters 108, 080402, doi: 10.1103/PhysRevLett.108.080402 (2012).
 35.
Chowdhury, A. N. & Somma, R. D. Quantum algorithms for Gibbs sampling and hittingtime estimation. arXiv:1603.02940 (2016).
 36.
Wocjan, P., Chiang, C.F., Nagaj, D. & Abeyesinghe, A. Quantum algorithm for approximating partition functions. Physical Review A 80, 022340, doi: 10.1103/PhysRevA.80.022340 (2009).
 37.
Boixo, S., Knill, E. & Somma, R. D. Quantum state preparation by phase randomization. Quantum Information & Computation 9, 833–855 (2009).
 38.
Richter, P. C. Quantum speedup of classical mixing processes. Physical Review A 76, 042306, doi: 10.1103/PhysRevA.76.042306 (2007).
 39.
Somma, R. D., Boixo, S., Barnum, H. & Knill, E. Quantum simulations of classical annealing processes. Physical Review Letters 101, 130504, doi: 10.1103/PhysRevLett.101.130504 (2008).
 40.
Tucci, R. R. Quantum Gibbs sampling using Szegedy operators. arXiv:0910.1647 (2009).
 41.
Terhal, B. M. & DiVincenzo, D. P. Problem of equilibration and the computation of correlation functions on a quantum computer. Physical Review A 61, 022301, doi: 10.1103/PhysRevA.61.022301 (2000).
 42.
Somma, R. D., Boixo, S. & Barnum, H. Quantum simulated annealing. arXiv:0712.1008 (2007).
 43.
Wocjan, P. & Abeyesinghe, A. Speedup via quantum sampling. Physical Review A 78, 1–8, doi: 10.1103/PhysRevA.78.042336 (2008).
 44.
Grover, L. K. A fast quantum mechanical algorithm for database search. In Proceedings of STOC96, 28th Annual Symposium on Theory of Computing, 212–219 (1996).
 45.
Long, G. L. Grover algorithm with zero theoretical failure rate. Physical Review A 64, 22307, doi: 10.1103/PhysRevA.64.022307 (2001).
 46.
Luis, A. & Peřina, J. Optimum phaseshift estimation and the quantum description of the phase difference. Physical Review A 54, 4564–4570, doi: 10.1103/PhysRevA.54.4564 (1996).
 47.
Brassard, G., Hoyer, P., Mosca, M. & Tapp, A. Quantum amplitude amplification and estimation. arXiv:quantph/0005055 (2000).
 48.
Kastoryano, M. J. & Brandão, F. G. S. L. Quantum Gibbs samplers: The commuting case. Communications in Mathematical Physics 344, 915–957, doi: 10.1007/s0022001626418 (2016).
 49.
Bilgin, E. & Boixo, S. Preparing thermal states of quantum systems by dimension reduction. Physical Review Letters 105, 170405, doi: 10.1103/PhysRevLett.105.170405 (2010).
 50.
Temme, K., Osborne, T. J., Vollbrecht, K. G., Poulin, D. & Verstraete, F. Quantum Metropolis sampling. Nature 471, 87–90, doi: 10.1038/nature09770 (2011).
 51.
Cook, S. A. The complexity of theoremproving procedures. In Proceedings of STOC71, 3rd Annual Symposium on Theory of Computing, 151–158 (1971).
 52.
Berry, D. W., Childs, A. M., Cleve, R., Kothari, R. & Somma, R. D. Simulating Hamiltonian dynamics with a truncated Taylor series. Physical Review Letters 114, 090502, doi: 10.1103/PhysRevLett.114.090502 (2015).
 53.
Wei, S. J. & Long, G. L. Duality quantum computer and the efficient quantum simulations. Quantum Information Processing 15, 1189–1212, doi: 10.1007/s1112801612636 (2016).
 54.
Long, G. L. General quantum interference principle and duality computer. Communications in Theoretical Physics 45, 825–844, doi: 10.1088/02536102/45/5/013 (2006).
 55.
Long, G. L. Duality quantum computing and duality quantum information processing. International Journal of Theoretical Physics 50, 1305–1318, doi: 10.1007/s107730100603z (2011).
 56.
Wei, S.J., Ruan, D. & Long, G.L. Duality quantum algorithm efficiently simulates open quantum systems. Scientific Reports 6, 30727, doi: 10.1038/srep30727 (2016).
 57.
Sarkhel, S., Venugopal, D., Singla, P. & Gogate, V. Lifted MAP inference for Markov logic networks. In Proceedings of AISTATS14, 17th International Conference on Artificial Intelligence and Statistics, 859–867 (2014).
 58.
Gogate, V. & Domingos, P. Probabilistic theorem proving. In Proceedings of UAI11, 27th Conference on Uncertainty in Artificial Intelligence, 256–265 (2011).
 59.
Johnson, M. W. et al. Quantum annealing with manufactured spins. Nature 473, 194–198, doi: 10.1038/nature10012 (2011).
 60.
Boixo, S. et al. Evidence for quantum annealing with more than one hundred qubits. Nature Physics 10, 218–224, doi: 10.1038/nphys2900 (2014).
 61.
O’Gorman, B. A., PerdomoOrtiz, A., Babbush, R., AspuruGuzik, A. & Smelyanskiy, V. Bayesian network structure learning using quantum annealing. European Physics Journal Special Topics 224, 163–188, doi: 10.1140/epjst/e2015023499 (2015).
 62.
Zaribafiyan, A., Marchand, D. J. J. & Rezaei, S. S. C. Systematic and deterministic graphminor embedding for Cartesian products of graphs. arXiv:1602.04274 (2016).
 63.
Benedetti, M., RealpeGómez, J., Biswas, R. & PerdomoOrtiz, A. Quantumassisted learning of graphical models with arbitrary pairwise connectivity. arXiv:1609.02542 (2016).
 64.
Lechner, W., Hauke, P. & Zoller, P. A quantum annealing architecture with alltoall connectivity from local interactions. Science Advances 1, e1500838–e1500838, doi: 10.1126/sciadv.1500838 (2015).
Acknowledgements
We would like to thank the anonymous referee 1 for suggesting to us refs 45, and 53,54,55,56 and the anonymous referee 2 for suggesting to us refs 7–11. P.W. and C.G. acknowledge financial support from the European Research Council (CoG QITBOX and AdG OSYRIS), the Axa Chair in Quantum Information Science, Spanish MINECO (FOQUS FIS201346768, QIBEQI FIS201680773P and Severo Ochoa Grant No. SEV20150522), Fundació Privada Cellex, and Generalitat de Catalunya (Grant No. SGR 874 and 875, and CERCA Programme). C.G. acknowledges support by the European Union’s Marie SkłodowskaCurie Individual Fellowships (IFEF) programme under GA: 700140.
Author information
Affiliations
ICFOThe Institute of Photonic Sciences, 08860 Castelldefels (Barcelona), Spain
 Peter Wittek
 & Christian Gogolin
University of Borås, 50190 Borås, Sweden
 Peter Wittek
Authors
Search for Peter Wittek in:
Search for Christian Gogolin in:
Contributions
P.W. and C.G. have contributed equally and wrote the main manuscript text. P.W. prepared Fig. 1. All authors reviewed the manuscript.
Competing interests
The authors declare no competing financial interests.
Corresponding author
Correspondence to Peter Wittek.
Rights and permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Further reading

1.
Nature (2017)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.