-
Explainable Natural Language Processing with Matrix Product States
Authors:
Jirawat Tangpanitanon,
Chanatip Mangkang,
Pradeep Bhadola,
Yuichiro Minato,
Dimitris G. Angelakis,
Thiparat Chotibut
Abstract:
Despite empirical successes of recurrent neural networks (RNNs) in natural language processing (NLP), theoretical understanding of RNNs is still limited due to intrinsically complex non-linear computations. We systematically analyze RNNs' behaviors in a ubiquitous NLP task, the sentiment analysis of movie reviews, via the map** between a class of RNNs called recurrent arithmetic circuits (RACs)…
▽ More
Despite empirical successes of recurrent neural networks (RNNs) in natural language processing (NLP), theoretical understanding of RNNs is still limited due to intrinsically complex non-linear computations. We systematically analyze RNNs' behaviors in a ubiquitous NLP task, the sentiment analysis of movie reviews, via the map** between a class of RNNs called recurrent arithmetic circuits (RACs) and a matrix product state (MPS). Using the von-Neumann entanglement entropy (EE) as a proxy for information propagation, we show that single-layer RACs possess a maximum information propagation capacity, reflected by the saturation of the EE. Enlarging the bond dimension beyond the EE saturation threshold does not increase model prediction accuracies, so a minimal model that best estimates the data statistics can be inferred. Although the saturated EE is smaller than the maximum EE allowed by the area law, our minimal model still achieves ~99% training accuracies in realistic sentiment analysis data sets. Thus, low EE is not a warrant against the adoption of single-layer RACs for NLP. Contrary to a common belief that long-range information propagation is the main source of RNNs' successes, we show that single-layer RACs harness high expressiveness from the subtle interplay between the information propagation and the word vector embeddings. Our work sheds light on the phenomenology of learning in RACs, and more generally on the explainability of RNNs for NLP, using tools from many-body quantum physics.
△ Less
Submitted 6 April, 2022; v1 submitted 16 December, 2021;
originally announced December 2021.
-
Hybrid Quantum-Classical Algorithms for Loan Collection Optimization with Loan Loss Provisions
Authors:
Jirawat Tangpanitanon,
Jirawat Saiphet,
Pantita Palittapongarnpim,
Poompong Chaiwongkhot,
Pinn Prugsanapan,
Nuntanut Raksasri,
Yarnvith Raksri,
Pairash Thajchayapong,
Thiparat Chotibut
Abstract:
Banks are required to set aside funds in their income statement, known as a loan loss provision (LLP), to account for potential loan defaults and expenses. By treating the LLP as a global constraint, we propose a hybrid quantum-classical algorithm to solve Quadratic Constrained Binary Optimization (QCBO) models for loan collection optimization. The objective is to find a set of optimal loan collec…
▽ More
Banks are required to set aside funds in their income statement, known as a loan loss provision (LLP), to account for potential loan defaults and expenses. By treating the LLP as a global constraint, we propose a hybrid quantum-classical algorithm to solve Quadratic Constrained Binary Optimization (QCBO) models for loan collection optimization. The objective is to find a set of optimal loan collection actions that maximizes the expected net profit presented to the bank as well as the financial welfare in the financial network of loanees, while kee** the LLP at its minimum. Our algorithm consists of three parts: a classical divide-and-conquer algorithm to enable a large-scale optimization, a quantum alternating operator ansatz (QAOA) algorithm to maximize the objective function, and a classical sampling algorithm to handle the LLP. We apply the algorithm to a real-world data set with 600 loanees and 5 possible collection actions. The QAOA is performed using up to 35 qubits on a classical computer. We show that the presence of the QAOA can improve the expected net profit by approximately $70\%$, compared to when the QAOA is absent from the hybrid algorithm. Our work illustrates the use of near-term quantum devices to tackle real-world optimization problems.
△ Less
Submitted 27 January, 2022; v1 submitted 29 October, 2021;
originally announced October 2021.
-
Quantum supremacy and quantum phase transitions
Authors:
Supanut Thanasilp,
Jirawat Tangpanitanon,
Marc-Antoine Lemonde,
Ninnat Dangniam,
Dimitris G. Angelakis
Abstract:
Demonstrating the ability of existing quantum platforms to perform certain computational tasks intractable to classical computers represents a cornerstone in quantum computing. Despite the growing number of such proposed "quantum supreme" tasks, it remains an important challenge to identify their direct applications. In this work, we describe how the approach proposed in Ref. [arXiv:2002.11946] fo…
▽ More
Demonstrating the ability of existing quantum platforms to perform certain computational tasks intractable to classical computers represents a cornerstone in quantum computing. Despite the growing number of such proposed "quantum supreme" tasks, it remains an important challenge to identify their direct applications. In this work, we describe how the approach proposed in Ref. [arXiv:2002.11946] for demonstrating quantum supremacy in generic driven analog many-body systems, such as those found in cold atom and ion setups, can be extended to explore dynamical quantum phase transitions. We show how key quantum supremacy signatures, such as the distance between the output distribution and the expected Porter Thomas distribution at the supremacy regime, can be used as effective order parameters. We apply this approach to a periodically driven disordered 1D Ising model and show that we can accurately capture the transition between the driven thermalized and many-body localized phases. This approach also captures the transition towards the Floquet prethermalized regime for high-frequency driving. Revisiting quantum phases of matter under the light of the recent discussions about quantum supremacy draws a link between complexity theory and analog many-body systems.
△ Less
Submitted 11 December, 2020;
originally announced December 2020.
-
Qubit-efficient encoding schemes for binary optimisation problems
Authors:
Benjamin Tan,
Marc-Antoine Lemonde,
Supanut Thanasilp,
Jirawat Tangpanitanon,
Dimitris G. Angelakis
Abstract:
We propose and analyze a set of variational quantum algorithms for solving quadratic unconstrained binary optimization problems where a problem consisting of $n_c$ classical variables can be implemented on $\mathcal O(\log n_c)$ number of qubits. The underlying encoding scheme allows for a systematic increase in correlations among the classical variables captured by a variational quantum state by…
▽ More
We propose and analyze a set of variational quantum algorithms for solving quadratic unconstrained binary optimization problems where a problem consisting of $n_c$ classical variables can be implemented on $\mathcal O(\log n_c)$ number of qubits. The underlying encoding scheme allows for a systematic increase in correlations among the classical variables captured by a variational quantum state by progressively increasing the number of qubits involved. We first examine the simplest limit where all correlations are neglected, i.e. when the quantum state can only describe statistically independent classical variables. We apply this minimal encoding to find approximate solutions of a general problem instance comprised of 64 classical variables using 7 qubits. Next, we show how two-body correlations between the classical variables can be incorporated in the variational quantum state and how it can improve the quality of the approximate solutions. We give an example by solving a 42-variable Max-Cut problem using only 8 qubits where we exploit the specific topology of the problem. We analyze whether these cases can be optimized efficiently given the limited resources available in state-of-the-art quantum platforms. Lastly, we present the general framework for extending the expressibility of the probability distribution to any multi-body correlations.
△ Less
Submitted 21 April, 2021; v1 submitted 3 July, 2020;
originally announced July 2020.
-
Expressibility and trainability of parameterized analog quantum systems for machine learning applications
Authors:
Jirawat Tangpanitanon,
Supanut Thanasilp,
Ninnat Dangniam,
Marc-Antoine Lemonde,
Dimitris G. Angelakis
Abstract:
Parameterized quantum evolution is the main ingredient in variational quantum algorithms for near-term quantum devices. In digital quantum computing, it has been shown that random parameterized quantum circuits are able to express complex distributions intractable by a classical computer, leading to the demonstration of quantum supremacy. However, their chaotic nature makes parameter optimization…
▽ More
Parameterized quantum evolution is the main ingredient in variational quantum algorithms for near-term quantum devices. In digital quantum computing, it has been shown that random parameterized quantum circuits are able to express complex distributions intractable by a classical computer, leading to the demonstration of quantum supremacy. However, their chaotic nature makes parameter optimization challenging in variational approaches. Evidence of similar classically-intractable expressibility has been recently demonstrated in analog quantum computing with driven many-body systems. A thorough investigation of trainability of such analog systems is yet to be performed. In this work, we investigate how the interplay between external driving and disorder in the system dictates the trainability and expressibility of interacting quantum systems. We show that if the system thermalizes, the training fails at the expense of the a large expressibility, while the opposite happens when the system enters the many-body localized (MBL) phase. From this observation, we devise a protocol using quenched MBL dynamics which allows accurate trainability while kee** the overall dynamics in the quantum supremacy regime. Our work shows the fundamental connection between quantum many-body physics and its application in machine learning. We conclude our work with an example application in generative modeling employing a well studied analog many-body model of a driven Ising spin chain. Our approach can be implemented with a variety of available quantum platforms including cold ions, atoms and superconducting circuits
△ Less
Submitted 22 May, 2020;
originally announced May 2020.
-
Quantum supremacy in driven quantum many-body systems
Authors:
Jirawat Tangpanitanon,
Supanut Thanasilp,
Marc-Antoine Lemonde,
Ninnat Dangiam,
Dimitris G. Angelakis
Abstract:
A crucial milestone in the field of quantum simulation and computation is to demonstrate that a quantum device can compute certain tasks that are impossible to reproduce by a classical computer with any reasonable resources. Such a demonstration is referred to as quantum supremacy. One of the most important questions is to identify setups that exhibit quantum supremacy and can be implemented with…
▽ More
A crucial milestone in the field of quantum simulation and computation is to demonstrate that a quantum device can compute certain tasks that are impossible to reproduce by a classical computer with any reasonable resources. Such a demonstration is referred to as quantum supremacy. One of the most important questions is to identify setups that exhibit quantum supremacy and can be implemented with current quantum technology. The two standard candidates are boson sampling and random quantum circuits. Here, we show that quantum supremacy can be obtained in generic periodically-driven quantum many-body systems. Our analysis is based on the eigenstate thermalization hypothesis and strongly-held conjectures in complexity theory. To illustrate our work, We give examples of simple disordered Ising chains driven by global magnetic fields and Bose-Hubbard chains with modulated hop**s. Our proposal opens the way for a large class of quantum platforms to demonstrate and benchmark quantum supremacy.
△ Less
Submitted 23 March, 2021; v1 submitted 27 February, 2020;
originally announced February 2020.
-
Many-Body Physics and Quantum Simulations with Strongly Interacting Photons
Authors:
Jirawat Tangpanitanon,
Dimitris G. Angelakis
Abstract:
Simulating quantum many-body systems on a classical computer generally requires a computational cost that grows exponential with the number of particles. This computational complexity has been the main obstacle to understanding various fundamental emergent phenomena in condensed matters such as high-Tc superconductivity and the fractional quantum-Hall effect. The difficulty arises because even the…
▽ More
Simulating quantum many-body systems on a classical computer generally requires a computational cost that grows exponential with the number of particles. This computational complexity has been the main obstacle to understanding various fundamental emergent phenomena in condensed matters such as high-Tc superconductivity and the fractional quantum-Hall effect. The difficulty arises because even the simplest models that are proposed to capture those phenomena cannot be simulated on a classical computer. Recognizing this problem in 1981, Richard Feynman envisioned a quantum simulator, an entirely new type of machine that exploits quantum superposition and operates by individually manipulating its constituting quantum particles and their interactions. Recent advances in various experimental platforms from cold atoms in optical lattices, trapped ions, to solid-state systems have brought the idea of Feynman to the realm of reality. Among those, interacting photons in superconducting circuits has been one of the promising platforms thanks to their local controllability and long coherence times. Early theoretical proposals have shown possibilities to realize quantum many-body phenomena of light using coupled cavity arrays such as Mott to superfluid transitions and fractional quantum Hall states. Start-of-the-art experiments include realization of interacting chiral edge states and stroboscopic signatures of localization of interacting photons in a three-site and a nine-site superconducting circuit, respectively. Interacting photons also serve as a natural platform to simulate driven-dissipative quantum many-body phenomena. A 72-site superconducting circuit has also recently been fabricated to study a dissipative phase transition of light.
△ Less
Submitted 11 July, 2019;
originally announced July 2019.
-
Quantum supremacy with analog quantum processors for material science and machine learning
Authors:
Jirawat Tangpanitanon,
Supanut Thanasilp,
Marc-Antoine Lemonde,
Dimitris G. Angelakis
Abstract:
Quantum supremacy is the ability of quantum processors to outperform classical computers at certain tasks. In digital random quantum circuit approaches for supremacy, the output distribution produced is described by the Porter-Thomas (PT) distribution. In this regime, the system uniformly explores its entire Hilbert space, which makes simulating such quantum dynamics with classical computational r…
▽ More
Quantum supremacy is the ability of quantum processors to outperform classical computers at certain tasks. In digital random quantum circuit approaches for supremacy, the output distribution produced is described by the Porter-Thomas (PT) distribution. In this regime, the system uniformly explores its entire Hilbert space, which makes simulating such quantum dynamics with classical computational resources impossible for large systems. However, the latter has no direct application so far in solving a specific problem. In this work, we show that the same sampling complexity can be achieved from driven analog quantum processors, with less stringent requirements for coherence and control. More importantly, we discuss how to apply this approach to solve problems in quantum simulations of phases of matter and machine learning. Specifically, we consider a simple quantum spin chain with nearest-neighbor interactions driven by a global magnetic field. We show how quantum supremacy is achieved as a consequence of the thermalization due to the interplay between the disorder and the driven many-body dynamics. We analyze how the achieved PT distribution can be used as an accessible reference distribution to probe the many-body localization (MBL) phase transition. In the second part of our work, we show how our setup can be used for generative modeling machine learning tasks. We propose a novel variational hybrid quantum-classical approach, exploiting the system's inherent tunable MBL dynamics, to train the device to learn distributions of complex classical data. The performance of our training protocol depends solely on the phase that the quantum system is in, which makes fine-tuning of local parameters not necessary. The protocol is implementable in a range of driven quantum many-body systems, compatible with noisy intermediate-scale quantum devices.
△ Less
Submitted 10 June, 2019;
originally announced June 2019.
-
Dynamical quantum phase transitions and non-Markovian dynamics
Authors:
Thi Ha Kyaw,
Victor M. Bastidas,
Jirawat Tangpanitanon,
Guillermo Romero,
Leong-Chuan Kwek
Abstract:
In the context of closed quantum systems, when a system prepared in its ground state undergoes a sudden quench, the resulting Loschmidt echo can exhibit zeros, resembling the Fisher zeros in the theory of classical equilibrium phase transitions. These zeros lead to nonanalytical behavior of the corresponding rate function, which is referred to as \textit{dynamical quantum phase transitions} (DQPTs…
▽ More
In the context of closed quantum systems, when a system prepared in its ground state undergoes a sudden quench, the resulting Loschmidt echo can exhibit zeros, resembling the Fisher zeros in the theory of classical equilibrium phase transitions. These zeros lead to nonanalytical behavior of the corresponding rate function, which is referred to as \textit{dynamical quantum phase transitions} (DQPTs). In this work, we investigate DQPTs in the context of open quantum systems that are coupled to both Markovian and non-Markovian dephasing baths via a conserved quantity. The general framework is corroborated by studying the non-equilibrium dynamics of a transverse-field Ising ring. We show the robustness of DQPT signatures under the action of both engineered dephasing baths, independently on how strongly they couple to the quantum system. Our theory provides insight on the effect of non-Markovian environments on DQPTs.
△ Less
Submitted 22 March, 2019; v1 submitted 12 November, 2018;
originally announced November 2018.
-
Strongly correlated photon transport in nonlinear photonic lattice with disorder: Probing signatures of the localization transition
Authors:
Tian Feng See,
V. M. Bastidas,
Jirawat Tangpanitanon,
Dimitris G. Angelakis
Abstract:
We study the transport of few-photon states in an open disordered nonlinear photonic lattice. More specifically, we consider a waveguide quantum electrodynamics (QED) setup where photons are scattered from a chain of nonlinear resonators with onsite Bose-Hubbard interaction in the presence of an incommensurate potential. Applying our recently developed diagrammatic technique that represents scatte…
▽ More
We study the transport of few-photon states in an open disordered nonlinear photonic lattice. More specifically, we consider a waveguide quantum electrodynamics (QED) setup where photons are scattered from a chain of nonlinear resonators with onsite Bose-Hubbard interaction in the presence of an incommensurate potential. Applying our recently developed diagrammatic technique that represents scattering matrix (S-matrix) with scattering diagrams and associated propagators, we compute the two-photon transmission probability and show that it carries signatures of the underlying many-body localization transition of the system. We compare the calculated probability to the participation ratio of the eigenstates and find close agreement for a range of interaction strengths. We analyze the robustness of the transmission signatures against local dissipation and briefly discuss possible implementation using current superconducting circuit technology.
△ Less
Submitted 19 March, 2019; v1 submitted 20 July, 2018;
originally announced July 2018.
-
Discrete time crystal in globally driven interacting quantum systems without disorder
Authors:
Wing Chi Yu,
Jirawat Tangpanitanon,
Alexander W. Glaetzle,
Dieter Jaksch,
Dimitris G. Angelakis
Abstract:
Time crystals in periodically driven systems have initially been studied assuming either the ability to quench the Hamiltonian between different many-body regimes, the presence of disorder or long-range interactions. Here we propose the simplest scheme to observe discrete time crystal dynamics in a one-dimensional driven quantum system of the Ising type with short-range interactions and no disorde…
▽ More
Time crystals in periodically driven systems have initially been studied assuming either the ability to quench the Hamiltonian between different many-body regimes, the presence of disorder or long-range interactions. Here we propose the simplest scheme to observe discrete time crystal dynamics in a one-dimensional driven quantum system of the Ising type with short-range interactions and no disorder. The system is subject only to a periodic kick by a global magnetic field, and no extra Hamiltonian quenching is performed. We analyze the emerging time crystal stabilizing mechanisms via extensive numerics as well as using an analytic approach based on an off-resonant transition model. Due to the simplicity of the driven Ising model, our proposal can be implemented with current experimental platforms including trapped ions, Rydberg atoms, and superconducting circuits.
△ Less
Submitted 26 March, 2019; v1 submitted 20 July, 2018;
originally announced July 2018.
-
Hidden Order in Quantum Many-body Dynamics of Driven-Dissipative Nonlinear Photonic Lattices
Authors:
Jirawat Tangpanitanon,
Stephen R. Clark,
V. M. Bastidas,
Rosario Fazio,
Dieter Jaksch,
Dimitris G. Angelakis
Abstract:
We study the dynamics of nonlinear photonic lattices driven by two-photon parametric processes. By means of matrix-product-state based calculations, we show that a quantum many-body state with long-range hidden order can be generated from the vacuum. This order resembles that characterizing the Haldane insulator. A possible explanation highlighting the role of the symmetry of the drive, and the ef…
▽ More
We study the dynamics of nonlinear photonic lattices driven by two-photon parametric processes. By means of matrix-product-state based calculations, we show that a quantum many-body state with long-range hidden order can be generated from the vacuum. This order resembles that characterizing the Haldane insulator. A possible explanation highlighting the role of the symmetry of the drive, and the effect of photon loss are discussed. An implementation based in superconducting circuits is proposed and analyzed
△ Less
Submitted 28 June, 2018;
originally announced June 2018.
-
Spectral signatures of many-body localization with interacting photons
Authors:
P. Roushan,
C. Neill,
J. Tangpanitanon,
V. M. Bastidas,
A. Megrant,
R. Barends,
Y. Chen,
Z. Chen,
B. Chiaro,
A. Dunsworth,
A. Fowler,
B. Foxen,
M. Giustina,
E. Jeffrey,
J. Kelly,
E. Lucero,
J. Mutus,
M. Neeley,
C. Quintana,
D. Sank,
A. Vainsencher,
J. Wenner,
T. White,
H. Neven,
D. G. Angelakis
, et al. (1 additional authors not shown)
Abstract:
Statistical mechanics is founded on the assumption that a system can reach thermal equilibrium, regardless of the starting state. Interactions between particles facilitate thermalization, but, can interacting systems always equilibrate regardless of parameter values\,? The energy spectrum of a system can answer this question and reveal the nature of the underlying phases. However, most experimenta…
▽ More
Statistical mechanics is founded on the assumption that a system can reach thermal equilibrium, regardless of the starting state. Interactions between particles facilitate thermalization, but, can interacting systems always equilibrate regardless of parameter values\,? The energy spectrum of a system can answer this question and reveal the nature of the underlying phases. However, most experimental techniques only indirectly probe the many-body energy spectrum. Using a chain of nine superconducting qubits, we implement a novel technique for directly resolving the energy levels of interacting photons. We benchmark this method by capturing the intricate energy spectrum predicted for 2D electrons in a magnetic field, the Hofstadter butterfly. By increasing disorder, the spatial extent of energy eigenstates at the edge of the energy band shrink, suggesting the formation of a mobility edge. At strong disorder, the energy levels cease to repel one another and their statistics approaches a Poisson distribution - the hallmark of transition from the thermalized to the many-body localized phase. Our work introduces a new many-body spectroscopy technique to study quantum phases of matter.
△ Less
Submitted 20 December, 2017; v1 submitted 20 September, 2017;
originally announced September 2017.
-
Floquet stroboscopic divisibility in non-Markovian dynamics
Authors:
V. M. Bastidas,
Thi Ha Kyaw,
Jirawat Tangpanitanon,
Guillermo Romero,
Leong-Chuan Kwek,
Dimitris G. Angelakis
Abstract:
We provide a general discussion of the Liouvillian spectrum for a system coupled to a non-Markovian bath using Floquet theory. This approach is suitable when the system is described by a time-convolutionless master equation with time-periodic rates. Surprisingly, the periodic nature of rates allow us to have a stroboscopic divisible dynamical map at discrete times, which we refer to as Floquet str…
▽ More
We provide a general discussion of the Liouvillian spectrum for a system coupled to a non-Markovian bath using Floquet theory. This approach is suitable when the system is described by a time-convolutionless master equation with time-periodic rates. Surprisingly, the periodic nature of rates allow us to have a stroboscopic divisible dynamical map at discrete times, which we refer to as Floquet stroboscopic divisibility. We illustrate the general theory for a Schrödinger cat which is roaming inside a non-Markovian bath, and demonstrate the appearance of stroboscopic revival of the cat at later time after its death. Our theory may have profound implications in entropy production in non-equilibrium systems.
△ Less
Submitted 28 September, 2018; v1 submitted 14 July, 2017;
originally announced July 2017.
-
Topological pum** of photons in nonlinear resonator arrays
Authors:
Jirawat Tangpanitanon,
Victor M. Bastidas,
Sarah Al-Assam,
Pedram Roushan,
Dieter Jaksch,
Dimitris G. Angelakis
Abstract:
We show how to implement topological or Thouless pum** of interacting photons in one dimensional nonlinear resonator arrays, by simply modulating the frequency of the resonators periodically in space and time. The interplay between interactions and the adiabatic modulations enables robust transport of Fock states with few photons per site. We analyze the transport mechanism via an effective anal…
▽ More
We show how to implement topological or Thouless pum** of interacting photons in one dimensional nonlinear resonator arrays, by simply modulating the frequency of the resonators periodically in space and time. The interplay between interactions and the adiabatic modulations enables robust transport of Fock states with few photons per site. We analyze the transport mechanism via an effective analytic model and study its topological properties and its protection to noise. We conclude by a detailed study of an implementation with existing circuit QED architectures.
△ Less
Submitted 14 July, 2016;
originally announced July 2016.
