References
 (1) L.S. Bishop, S. Bravyi, A. Cross, J.M. Gambetta, and J. Smolin, Quantum volume, Technical report, IBM T.J. Watson, 2017.
 (2) A.I. Lvovsky and M.G. Raymer, Continuousvariable optical quantumstate tomography, Rev. Mod. Phys. 81, 299 (2009).
 (3) U. Leonhardt, Measuring the quantum state of light, (Cambridge University Press, Cambridge, 1997).
 (4) D. Barredo, S. de Leseleuc, V. Lienhard, T. Lahaye, and A. Browaeys, An atombyatom assembler of defectfree arbitrary twodimensional atomic arrays, Science 354, 1021 (2016).
 (5) M. Endres, H. Bernien, A. Keesling, H. Levine, E.R. Anschuetz, A. Krajenbrink, C. Senko, V. Vuletić, M. Greiner, and M.D. Lukin, Atombyatom assembly of defectfree onedimensional cold atom arrays, Science 354, 1024 (2016).
 (6) H. Bernien, S. Schwartz, A. Keesling, H. Levine, A. Omran, H. Pichler, S. Choi, A.S. Zibrov, M. Endres, M. Greiner, V. Vuletić, and M.D. Lukin, Probing manybody dynamics on a 51atom quantum simulator, Nature (London) 551, 579 (2017).
 (7) D. Barredo, V. Lienhard, S. de Léséleuc, T. Lahaye, and A. Browaeys, Synthetic threedimensional atomic structures assembled atom by atom, Nature (London) 561, 79 (2018).
 (8) J. Zhang, G. Pagano, P. W. Hess, A. Kyprianidis, P. Becker, H. Kaplan, A.V. Gorshkov, Z.X. Gong, and C. Monroe, Observation of a manybody dynamical phase transition with a 53qubit quantum simulator, Nature (London) 551, 601 (2017).
 (9) C. Neill, P. Roushan, K. Kechedzhi, S. Boixo, S. V. Isakov, V. Smelyanskiy, R. Barends, B. Burkett, Y. Chen, Z. Chen, B. Chiaro, A. Dunsworth, A. Fowler, B. Foxen, R.Graff, E. Jeffrey, J. Kelly, E. Lucero, A. Megrant, J. Mutus, M. Neeley, C. Quintana, D. Sank, A. Vainsencher, J. Wenner, T.C. White, H. Neven, and J.M. Martinis, A blueprint for demonstrating quantum supremacy with superconducting qubits, Science 360, 195 (2018).
 (10) U.L. Andersen, J.S. NeergaardNielsen, P. van Loock, and A. Furusawa, Hybrid discrete and continuousvariable quantum information, Nat. Phys. 11, 713 (2015).
 (11) G. Toth, W. Wieczorek, D. Gross, R. Krischek, C. Schwemmer, and H. Weinfurter, Permutationally invariant quantum tomography, Phys. Rev. Lett. 105, 250403 (2010).
 (12) D. Gross, Y.K. Liu, S.T. Flammia, S. Becker, and J. Eisert, Quantum state tomography via compressed sensing, Phys. Rev. Lett. 105, 150401 (2010).
 (13) M. Cramer, M.B. Plenio, S.T. Flammia, D. Gross, S.D. Bartlett, R. Somma, O. LandonCardinal, D. Poulin, and Y.K. Liu, Efficient quantum state tomography, Nat. Commun. 1, 149 (2010).
 (14) B.P. Lanyon, C. Maier, M. Holzpfel, T. Baumgratz, C. Hempel, P. Jurcevic, I. Dhand, A.S. Buyskikh, A.J. Daley, M. Cramer, M.B. Plenio, R. Blatt, and C.F. Roos, Efficient tomography of a quantum manybody system, Nat. Phys. 13, 1158 (2017).
 (15) J. Carrasquilla, G. Torlai, R.G. Melko, and L. Aolita, Reconstructing quantum states with generative models, Nat. Mach. Intell. 1, 155 (2019).

(16)
G. Cybenko, Approximations by superpositions of sigmoidal functions,
Math. Control Signals Syst. 2, 303 (1989).  (17) G. Carleo and M. Troyer, Solving the quantum manybody problem with artificial neural networks, Science 355, 602 (2017).
 (18) R. Melko, G. Carleo, J. Carrasquilla, and J.I. Cirac, Restricted Boltzmann machines in quantum physics, Nat. Phys. (2019).
 (19) G. Torlai, G. Mazzola, J. Carrasquilla, M. Troyer, R. Melko, and G. Carleo, Neuralnetwork quantum state tomography, Nat. Phys. 14, 447 (2017).
 (20) R. Xia and S. Kais, Quantum machine learning for electronic structure calculations, Nat. Commun. 9, 4195 (2018).
 (21) J. Chen, S. Cheng, H. Xie, L. Wang, and T. Xiang, Equivalence of restricted Boltzmann machines and tensor network states, Phys. Rev. B 97, 085104 (2018).
 (22) S. Lu, X. Gao, and L.M. Duan, Efficient representation of topologically ordered states with restricted Boltzmann machines, Phys. Rev. B 99, 155136 (2019).
 (23) M. Schuld, I. Sinayskiy, and F. Petruccione, Viewpoint: Neural networks take on open quantum systems, Physics 12, 74 (2019).
 (24) N. Yoshioka and R. Hamazaki, Constructing neural stationary states for open quantum manybody systems, Phys. Rev. B 99, 214306 (2019).
 (25) A. Nagy and V. Savona, Variational quantum Monte Carlo method with a neuralnetwork ansatz for open quantum systems, Phys. Rev. Lett. 122, 250501 (2019).
 (26) M.J. Hartmann and G. Carleo, Neuralnetwork approach to dissipative quantum manybody dynamics, Phys. Rev. Lett. 122, 250502 (2019).
 (27) F. Vicentini, A. Biella, N. Regnault, and C. Ciuti, Variational neuralnetwork ansatz for steady states in open quantum systems, Phys. Rev. Lett. 122, 250503 (2019).
 (28) G. Torlai and R.G. Melko, Latent space purification via neural density operators, Phys. Rev. Lett. 120, 240503 (2018).
 (29) G. Torlai, B. Timar, E.P.L. van Nieuwenburg, H. Levine, A. Omran, A. Keesling, H. Bernien, M. Greiner, V. Vuletić, M.D. Lukin, R.G. Melko, and M. Endres, Integrating neural networks with a quantum simulator for state reconstruction, arXiv:1904.08441.
 (30) A.M. Palmieri, E. Kovlakov, F. Bianchi, D. Yudin, S. Straupe, J. Biamonte, and S. Kulik, Experimental neural network enhanced quantum tomography, arXiv:1904.05902.
 (31) A.I. Lvovsky, Iterative maximumlikelihood reconstruction in quantum homodyne tomography, J. Opt. B 6, S556 (2004).
 (32) H. Larochelle and R. Bengio, Classification using discriminative restricted Boltzmann machines, in Proceedings of the 25th International Conference on Machine Learning, New York, USA (ACM, New York, 2008), pp. 536543.
 (33) H. Hu, L. Gao, and Q. Ma, Deep restricted Boltzmann networks, arXiv:1611.07917.
 (34) Z. Hradil, J. Řeháček, J. Fiurášek, M. Ježek, MaximumLikelihood methods in quantum mechanics, Lect. Notes Phys. 649, 59 (2004).
 (35) D.V. Sychev, A.E. Ulanov, A.A. Pushkina, M.W. Richards, I.A. Fedorov, and A.I. Lvovsky, Enlargement of optical Schrödinger’s cat states, Nat. Photon. 11, 379 (2017).
 (36) E. Bimbard, N. Jain, A. MacRae and A.I. Lvovsky, Quantumoptical state engineering up to the twophoton level, Nat. Photon. 4, 243 (2010).
 (37) J. Yoshikawa, S. Yokoyama, T. Kaji, C. Sornphiphatphong, Y. Shiozawa, K. Makino, and A. Furusawa, Generation of onemillionmode continuousvariable cluster state by unlimited timedomain multiplexing, APL Photonics 1, 060801 (2016).
 (38) Y. Cai, J. Roslund, G. Ferrini, F. Arzani, X. Xu, C. Fabre, and N. Treps, Multimode entanglement in reconfigurable graph states using optical frequency combs, Nat. Commun. 8, 15645 (2017).
 (39) K. Hammerer, A.S. Sørensen, and E.S. Polzik, Quantum interface between light and atomic ensembles, Rev. Mod. Phys. 82, 1041 (2010).
 (40) M. Aspelmeyer, T.J. Kippenberg, and F. Marquardt, Cavity Optomechanics (Springer Berlin Heidelberg, Berlin, 2014).

(41)
M.A. CarreiraPerpinan and G.E. Hinton, On contrastive divergence learning, Artificial Intelligence and Statistics
2005, 17 (2005).  (42) S. Geman and D. Geman. Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images, IEEE Trans. Pattern Anal. Mach. Intell. 6, 721 (1984).
 (43) X. Gao and L.M. Duan, Efficient representation of quantum manybody states with deep neural networks, Nat. Commun. 8, 662 (2017).
I Methods
Pure states.
Here we present the details of training our RBM. The neural network parametrization for the wavefunction is defined by Eqs. (4) and (5). We introduce additional notation. First, because there is onetoone correspondence between visible layer configurations and Fock states , as discussed in the main text, we will use the symbol to denote both these objects. Second, we denote the unnormalized Boltzmann probability
(8) 
with , for the amplitude RBM, and the analogous quantity
(9) 
for the phase RBM. We remind the reader that the letters and denote the respective parameter sets of these RBMs.
By plugging the expression (4) into the loglikelihood function (6) and using that the fact that overlap between the number and quadrature eigenstates corresponding to the phase is just the
’s harmonic oscillator eigenfunction (HermiteGaussian polynomial)
with a phase factor,(10) 
we obtain the following expression:
(11)  
where , the summation with index is over all quadrature measurements and is the number of measurements.
For the network training, we evaluate the gradients of the above loglikelihood over the neural net parameters and as follows:
(12a)  
(12b) 
where we defined
(13a)  
(13b) 
with and . Ascending by these gradients, we can maximize the loglikelihood (11). Both RBMs are trained simultaneously.
We note that the above gradients contain exhaustive summation over possible configutations of the visible and hidden layers of both RBMs. In the present work, we are able to compute this sum directly since the number of RBM units is relatively small. However, in the case of high Hilbert space dimension, Boltzmann sampling using an annealing device or algorithm will be required.
Mixed states.
As discussed in the main text, see Eq. (7), we treat the mixed state to be reconstructed as a partial state of a pure state in a tensor product Hilbert space with the dimension . We decompose this state in the Fock basis and apply the same parametrization as in the previous subsection:
(14)  
(15) 
The partial trace of this state over the environment is as follows:
(16)  
The loglikelihood (6) is then given by
(17)  
where the summation indices run over the truncated Fock basis, over all quadrature measurements, and
(18)  
We note that the expression (17) is very similar to the pure state case (11), but requires two additional summations over the truncated Fock basis. The loglikelihood (17) gradients over and read similarly to those for the pure state (12), but with the parameters (13) redefined as follows:
(19a)  
(19b) 
The remainder of the treatment replicates that for pure states.
Efficiency correction.
To correct for an imperfect homodyne detector efficiency in our neural net approach, we model it as a perfect detector preceded by beam splitter of transmission Lvovsky2004 , which changes the quantum state by means of generalized Bernoulli transformation to a new state according to
(20) 
where . Now we can repeat the above procedure for the mixed state (purification) Ansatz, with the only difference that we use instead of to calculate the loglikelihood (6).
Comments
There are no comments yet.