Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:1711.09442 · Scientific Reports

Quantum Artificial Life in an IBM Quantum Computer

Unai Alvarez-Rodriguez, Mikel Sanz, Lucas Lamata, E. Solano

Published Sep 28, 2018• 90 citations
arXivHTMLPDF

Abstract

We present the first experimental realization of a quantum artificial life algorithm in a quantum computer. The quantum biomimetic protocol encodes tailored quantum behaviors belonging to living systems, namely, self-replication, mutation, interaction between individuals, and death, into the cloud quantum computer IBM ibmqx4. In this experiment, entanglement spreads throughout generations of individuals, where genuine quantum information features are inherited through genealogical networks. As a pioneering proof-of-principle, experimental data fits the ideal model with accuracy. Thereafter, these and other models of quantum artificial life, for which no classical device may predict its quantum supremacy evolution, can be further explored in novel generations of quantum computers. Quantum biomimetics, quantum machine learning, and quantum artificial intelligence will move forward hand in hand through more elaborate levels of quantum complexity.

DOI
10.1038/s41598-018-33125-3

Related papers

Quantum Artificial Life Takes Its First Step on an IBM Quantum Computer

inquantio

A research team based in Spain modeled the four ingredients of Darwinian evolution, self replication, mutation, interaction between individuals, and death, as a quantum circuit and ran it for the first time on IBM's cloud quantum computer, ibmqx4. Each individual was defined as two qubits, a genotype and a phenotype, and successive generations were linked by partial quantum cloning, a procedure that works around the no cloning theorem. The complete model combining all four ingredients reproduced the ideal probability distribution with a fidelity of 93.94%, though the researchers were explicit that this is a proof of principle, a simulation of the logic of life on quantum hardware rather than an account of how life began. [Quantum Biology Society] Where quantum biology has mostly gone looking for quantum mechanical phenomena inside living organisms, this study runs the argument in the opposite direction: it implements the logic of biological evolution inside the physical environment of a quantum computer. The authors describe their line of work as quantum biomimetics, the design of quantum algorithms that imitate biological processes. Researchers at the University of the Basque Country (UPV/EHU) in Bilbao, together with colleagues at several partner institutions, designed the elements of Darwinian natural selection and evolution (self replication, mutation, interaction, and death) as a quantum algorithm that works around the no cloning theorem, one of the founding principles of quantum mechanics. They then succeeded in running it on ibmqx4, the five qubit superconducting chip IBM makes available over the cloud, putting into practice a model the same group had proposed two years earlier. The results, published in Scientific Reports in 2018, stand as the first experimental implementation of a quantum artificial life algorithm on quantum hardware. ■ Two Qubits per Individual: Partial Quantum Cloning Sidesteps the No Cloning Theorem The smallest unit of life in this model, the individual, is built from two qubits, one holding the genotype and one the phenotype. The genotype carries the information that identifies the type of living unit and is handed on from generation to generation. The phenotype is set by the genetic information together with the interaction between the individual and its environment, and that state, along with the information it encodes, is degraded over the individual's lifetime, which is how aging and death enter the model. Self replication, the core of life, is implemented through partial quantum cloning. Quantum mechanics forbids making a perfect copy of an unknown quantum state in full, but it does permit the expectation value of a single chosen observable, here the Pauli Z observable σz, to be written into a blank qubit. Using controlled NOT (CNOT) gates, the researchers passed that particular piece of the genotype's information first into the genotype qubit of the next generation and then into its phenotype qubit, completing a system of heredity in the quantum world. ■ A Practical Compromise with the Hardware: Emulating Dissipation and Mutation Because the flexibility of current quantum hardware is limited, death and mutation had to be simulated by practical approximations. Death should be produced by leaving the phenotype qubit open to its environment so that its information gradually dissipates, but the device could not perform the conditional projection of the phenotype onto the |0⟩ state that this calls for. Instead the researchers applied a sequence of small rotations about the σy axis, one for each simulated time step, to imitate the running out of a lifetime. As they note, this replaces the dissipation with a discrete process and loses its exponential character. Mutation, likewise, was not implemented as a fully random rotation. The researchers fixed the mutation to a single bit flip (σx) and then took a statistical detour: circuits with and without a mutation were run different numbers of times and the data merged afterward so that the intended mutation rate came out of the aggregate. Interaction between individuals is a conditional operation that swaps phenotypes according to the genotypes, and it was built by combining two qubit gates including CNOT, SWAP, and the controlled square root of NOT. ■ The Results, and the Mutation Paradox The experiments proceeded by adding the elements of evolution one after another: interaction between individuals, self replication, dissipation into the environment, and mutation. For the complete model with all four in place, the measured probability distribution matched the ideal prediction with a fidelity of 93.94%. One result along the way was especially curious. When mutation, the element that injects randomness into the system, was included, the fidelity of the data went up rather than down. The researchers give a level headed account of it: mutations tend to homogenize the probability distribution, which is the same effect the errors already present in the experimental platform produce. They also remark on the intriguing parallel with the classical world, where mutation likewise helps a species adapt to changing environments. ■ What Is Actually Quantum Here: Entanglement That Remembers a Lineage The simpler statistical indicators can be reproduced by a classical computer model without much difficulty. Every indicator based on measurements in the σz basis, the researchers point out, can be matched by a classical probability distribution. So which part of this model is genuinely quantum? The answer shows up in the correlations. A partial quantum cloning operation entangles the qubits it involves, carrying the expectation value of σx in the original state into the joint value of σx⊗σx. In a simplified case without mutations or interactions between individuals, measuring σx⊗σx⊗σx⊗σx across the four qubits of two individuals gives α if the two are linked by a self replication event, and α² if they are not causally connected, where α is the expectation value of σx in the precursor genotype. The two histories are therefore distinguishable from the correlation alone. Had the precursor genotype been a classical incoherent mixture instead, that expectation value would be zero and nothing would propagate into these global correlations at all. In other words, without keeping a separate ledger of who descended from whom, as a classical model would have to, the entangled state of the individuals itself records and attests to the lineage. On the chip this signature survived only in part: the measured value of σx⊗σx⊗σx⊗σx was 0.22 against an ideal 0.56, a sizable error. It was nonetheless clearly different from the product of the correlations within each individual taken separately, which comes out at zero, and it is that difference which shows the two individuals to be causally related. ■ Significance and Limits The study is a well executed proof of principle that a microscopic quantum system can hold the grammar of Darwinian evolution consistently, and that it can be run on a real superconducting quantum chip. The paper is explicit, however, that it is not an observation of quantum effects at work in a living organism. The authors leave open the philosophical question of whether the origin of life is genuinely quantum mechanical, and they state that the quantum supremacy referred to in the abstract and conclusion was not achieved in this experiment but remains a target to be explored on larger quantum computers in the future. They also acknowledge that the part of the dynamics carried by σz, which encodes the lifetime and the predator prey character of an individual, is classical information that a simpler classical analogue could predict; what the classical version cannot supply is the entanglement linking the living units. In asking whether the logic of life can be recoded in the language of quantum mechanics, this work sets out another blueprint for where quantum algorithms and the biological sciences may converge. #QuantumBiology #QuantumComputer #IBMQuantumComputer #QuantumArtificialLife #ArtificialLife #Qubit #QuantumEntanglement #NoCloningTheorem #DarwinianEvolution https://www.nature.com/articles/s41598-018-33125-3

•Jul 21, 2026•0 citations

Artificial Life in Quantum Technologies

Unai Alvarez-Rodriguez, Mikel Sanz, Lucas Lamata +1 more

We develop a quantum information protocol that models the biological behaviours of individuals living in a natural selection scenario. The artificially engineered evolution of the quantum living units shows the fundamental features of life in a common environment, such as self-replication, mutation, interaction of individuals, and death. We propose how to mimic these bio-inspired features in a quantum-mechanical formalism, which allows for an experimental implementation achievable with current quantum platforms. This study paves the way for the realization of artificial life and embodied evolution with quantum technologies.

arXiv:1505.03775•Scientific Reports•Feb 8, 2016•30 citations

Quantum machine learning and quantum biomimetics: A perspective

Lucas Lamata

Abstract Quantum machine learning has emerged as an exciting and promising paradigm inside quantum technologies. It may permit, on the one hand, to carry out more efficient machine learning calculations by means of quantum devices, while, on the other hand, to employ machine learning techniques to better control quantum systems. Inside quantum machine learning, quantum reinforcement learning aims at developing ‘intelligent’ quantum agents that may interact with the outer world and adapt to it, with the strategy of achieving some final goal. Another paradigm inside quantum machine learning is that of quantum autoencoders, which may allow one for employing fewer resources in a quantum device via a training process. Moreover, the field of quantum biomimetics aims at establishing analogies between biological and quantum systems, to look for previously inadvertent connections that may enable useful applications. Two recent examples are the concepts of quantum artificial life, as well as of quantum memristors. In this Perspective, we give an overview of these topics, describing the related research carried out by the scientific community.

arXiv:2004.12076•Machine Learning Science and Technology•May 30, 2020•61 citations

Quantum Neural Machine Learning: Theory and Experiments

Carlos Pedro Gonçalves

Cloud-based access to quantum computers opens up the way for the empirical implementation of quantum artificial neural networks and for the future integration of quantum computation in different devices, using the cloud to access a quantum computer. The current work experimentally implements quantum artificial neural networks on IBM’s quantum computers, accessed via cloud. Examples are provided for the XOR Boolean function representation problem and decision under risk; in the last case, quantum object-oriented programming using IBM’s Qiskit Python library is employed to implement a form of quantum neural reinforcement learning applied to a classical decision under risk problem, showing how decision can be integrated into a quantum artificial intelligence system, where an artificial agent learns how to select an optimal action when facing a classical gamble. A final reflection is provided on quantum robotics and a future where robotic systems are connected to quantum computers via cloud, using quantum neural computation to learn to optimize tasks and act accordingly.

IntechOpen eBooks•Feb 12, 2019•13 citations

Quantum Machine Learning and Bioinspired Quantum Technologies

Lucas Lamata, Mikel Sanz, E. Solano

There is no doubt artificial intelligence and machine learning have been significantly modifying the scientific and technological landscape, as well as society at large, over the last decade. Search engines, autonomous cars, language and speech translators, as well as image classifiers are only some possibilities this discipline enables and there is certainly more to come in the near future. Significant research efforts have been dedicated in the direction of quantum computing, a topic that has been receiving ever-growing attention from governmental institutions, multinational companies, and academics. A scalable quantum computer could offer unprecedented computational capabilities, enabling a revolution in information processing and transmission. Among the various approaches, quantum machine learning has emerged as an utterly successful method to better control and understand quantum systems, benefitting from machine learning algorithms, while relying on quantum processors to accelerate machine learning tasks. When coupled with biomimetics, an area of science that aims at imitating biological systems in order to design more efficient artificial devices, quantum biomimetics arises as a novel field that bridges the gap between natural selection in bio systems and engineering processes in quantum devices with one ultimate goal: The replication of biological behaviors in quantum controllable systems, what we call “bioinspired quantum technologies”. We believe that this Special Issue will shed light on unexplored areas up to now, acting as the precursor for further exciting research in the field of bioinspired quantum technologies.

Advanced Quantum Technologies•Aug 1, 2019•7 citations

Augmenting Quantum Mechanics with Artificial Intelligence

Giacomo Torlai

The simulation of quantum matter with classical hardware plays a central role in the discovery and development of quantum many-body systems, with far-reaching implications in condensed matter physics and quantum technologies. In general, efficient and sophisticated algorithms are required to overcome the severe challenge posed by the exponential scaling of the Hilbert space of quantum systems. In contrast, hardware built with quantum bits of information are inherently capable of efficiently finding solutions of quantum many-body problems. While a universal and scalable quantum computer is still beyond the horizon, recent advances in qubit manufacturing and coherent control of synthetic quantum matter are leading to a new generation of intermediate scale quantum hardware. \n \nThe complexity underlying quantum many-body systems closely resembles the one encountered in many problems in the world of information and technology. In both contexts, the complexity stems from a large number of interacting degrees of freedom. A powerful strategy in the latter scenario is machine learning, a subfield of artificial intelligence where large amounts of data are used to extract relevant features and patterns. In particular, artificial neural networks have been demonstrated to be capable of discovering low-dimensional representations of complex objects from high-dimensional dataset, leading to the profound technological revolution we all witness in our daily life. \n \nIn this Thesis, we envision a new paradigm for scientific discovery in quantum physics. On the one hand, we have the essentially unlimited data generated with the increasing amount of highly controllable quantum hardware. On the other hand, we have a set of powerful algorithms that efficiently capture non-trivial correlations from high-dimensional data. Therefore, we fully embrace this data-driven approach to quantum mechanics, and anticipate new exciting possibilities in the field of quantum many-body physics and quantum information science. We revive a powerful stochastic neural network called a restricted Boltzmann machine, which slowly moved out of fashion after playing a central role in the machine learning revolution of the early 2010s. We introduce a neural-network representation of quantum states based on this generative model. We propose a set of algorithms to reconstruct unknown quantum states from measurement data and numerically demonstrate their potential, with important implications for current experiments. These include the reconstruction of experimentally inaccessible properties, such as entanglement, and diagnostics to determine sources of noise. Furthermore, we introduce a machine learning framework for quantum error correction, where a neural network learns the best decoding strategy directly from data. We expect that the full integration between quantum hardware and artificial intelligence will become the gold standard, and will drive the world into the era of fault-tolerant quantum computing and large-scale quantum simulations.

•Dec 4, 2018•4 citations
We present the first experimental realization of a quantum artificial life algorithm in a quantum computer. The quantum biomimetic protocol encodes tailored quantum behaviors belonging to living systems, namely, self-replication, mutation, interaction between individuals, and death, into the cloud quantum computer IBM ibmqx4. In this experiment, entanglement spreads throughout generations of individuals, where genuine quantum information features are inherited through genealogical networks. As a pioneering proof-of-principle, experimental data fits the ideal model with accuracy. Thereafter, these and other models of quantum artificial life, for which no classical device may predict its quantum supremacy evolution, can be further explored in novel generations of quantum computers. Quantum biomimetics, quantum machine learning, and quantum artificial intelligence will move forward hand in hand through more elaborate levels of quantum complexity.
A research team based in Spain modeled the four ingredients of Darwinian evolution, self replication, mutation, interaction between individuals, and death, as a quantum circuit and ran it for the first time on IBM's cloud quantum computer, ibmqx4. Each individual was defined as two qubits, a genotype and a phenotype, and successive generations were linked by partial quantum cloning, a procedure that works around the no cloning theorem. The complete model combining all four ingredients reproduced the ideal probability distribution with a fidelity of 93.94%, though the researchers were explicit that this is a proof of principle, a simulation of the logic of life on quantum hardware rather than an account of how life began. [Quantum Biology Society] Where quantum biology has mostly gone looking for quantum mechanical phenomena inside living organisms, this study runs the argument in the opposite direction: it implements the logic of biological evolution inside the physical environment of a quantum computer. The authors describe their line of work as quantum biomimetics, the design of quantum algorithms that imitate biological processes. Researchers at the University of the Basque Country (UPV/EHU) in Bilbao, together with colleagues at several partner institutions, designed the elements of Darwinian natural selection and evolution (self replication, mutation, interaction, and death) as a quantum algorithm that works around the no cloning theorem, one of the founding principles of quantum mechanics. They then succeeded in running it on ibmqx4, the five qubit superconducting chip IBM makes available over the cloud, putting into practice a model the same group had proposed two years earlier. The results, published in Scientific Reports in 2018, stand as the first experimental implementation of a quantum artificial life algorithm on quantum hardware. ■ Two Qubits per Individual: Partial Quantum Cloning Sidesteps the No Cloning Theorem The smallest unit of life in this model, the individual, is built from two qubits, one holding the genotype and one the phenotype. The genotype carries the information that identifies the type of living unit and is handed on from generation to generation. The phenotype is set by the genetic information together with the interaction between the individual and its environment, and that state, along with the information it encodes, is degraded over the individual's lifetime, which is how aging and death enter the model. Self replication, the core of life, is implemented through partial quantum cloning. Quantum mechanics forbids making a perfect copy of an unknown quantum state in full, but it does permit the expectation value of a single chosen observable, here the Pauli Z observable σz, to be written into a blank qubit. Using controlled NOT (CNOT) gates, the researchers passed that particular piece of the genotype's information first into the genotype qubit of the next generation and then into its phenotype qubit, completing a system of heredity in the quantum world. ■ A Practical Compromise with the Hardware: Emulating Dissipation and Mutation Because the flexibility of current quantum hardware is limited, death and mutation had to be simulated by practical approximations. Death should be produced by leaving the phenotype qubit open to its environment so that its information gradually dissipates, but the device could not perform the conditional projection of the phenotype onto the |0⟩ state that this calls for. Instead the researchers applied a sequence of small rotations about the σy axis, one for each simulated time step, to imitate the running out of a lifetime. As they note, this replaces the dissipation with a discrete process and loses its exponential character. Mutation, likewise, was not implemented as a fully random rotation. The researchers fixed the mutation to a single bit flip (σx) and then took a statistical detour: circuits with and without a mutation were run different numbers of times and the data merged afterward so that the intended mutation rate came out of the aggregate. Interaction between individuals is a conditional operation that swaps phenotypes according to the genotypes, and it was built by combining two qubit gates including CNOT, SWAP, and the controlled square root of NOT. ■ The Results, and the Mutation Paradox The experiments proceeded by adding the elements of evolution one after another: interaction between individuals, self replication, dissipation into the environment, and mutation. For the complete model with all four in place, the measured probability distribution matched the ideal prediction with a fidelity of 93.94%. One result along the way was especially curious. When mutation, the element that injects randomness into the system, was included, the fidelity of the data went up rather than down. The researchers give a level headed account of it: mutations tend to homogenize the probability distribution, which is the same effect the errors already present in the experimental platform produce. They also remark on the intriguing parallel with the classical world, where mutation likewise helps a species adapt to changing environments. ■ What Is Actually Quantum Here: Entanglement That Remembers a Lineage The simpler statistical indicators can be reproduced by a classical computer model without much difficulty. Every indicator based on measurements in the σz basis, the researchers point out, can be matched by a classical probability distribution. So which part of this model is genuinely quantum? The answer shows up in the correlations. A partial quantum cloning operation entangles the qubits it involves, carrying the expectation value of σx in the original state into the joint value of σx⊗σx. In a simplified case without mutations or interactions between individuals, measuring σx⊗σx⊗σx⊗σx across the four qubits of two individuals gives α if the two are linked by a self replication event, and α² if they are not causally connected, where α is the expectation value of σx in the precursor genotype. The two histories are therefore distinguishable from the correlation alone. Had the precursor genotype been a classical incoherent mixture instead, that expectation value would be zero and nothing would propagate into these global correlations at all. In other words, without keeping a separate ledger of who descended from whom, as a classical model would have to, the entangled state of the individuals itself records and attests to the lineage. On the chip this signature survived only in part: the measured value of σx⊗σx⊗σx⊗σx was 0.22 against an ideal 0.56, a sizable error. It was nonetheless clearly different from the product of the correlations within each individual taken separately, which comes out at zero, and it is that difference which shows the two individuals to be causally related. ■ Significance and Limits The study is a well executed proof of principle that a microscopic quantum system can hold the grammar of Darwinian evolution consistently, and that it can be run on a real superconducting quantum chip. The paper is explicit, however, that it is not an observation of quantum effects at work in a living organism. The authors leave open the philosophical question of whether the origin of life is genuinely quantum mechanical, and they state that the quantum supremacy referred to in the abstract and conclusion was not achieved in this experiment but remains a target to be explored on larger quantum computers in the future. They also acknowledge that the part of the dynamics carried by σz, which encodes the lifetime and the predator prey character of an individual, is classical information that a simpler classical analogue could predict; what the classical version cannot supply is the entanglement linking the living units. In asking whether the logic of life can be recoded in the language of quantum mechanics, this work sets out another blueprint for where quantum algorithms and the biological sciences may converge. #QuantumBiology #QuantumComputer #IBMQuantumComputer #QuantumArtificialLife #ArtificialLife #Qubit #QuantumEntanglement #NoCloningTheorem #DarwinianEvolution https://www.nature.com/articles/s41598-018-33125-3
We develop a quantum information protocol that models the biological behaviours of individuals living in a natural selection scenario. The artificially engineered evolution of the quantum living units shows the fundamental features of life in a common environment, such as self-replication, mutation, interaction of individuals, and death. We propose how to mimic these bio-inspired features in a quantum-mechanical formalism, which allows for an experimental implementation achievable with current quantum platforms. This study paves the way for the realization of artificial life and embodied evolution with quantum technologies.
Abstract Quantum machine learning has emerged as an exciting and promising paradigm inside quantum technologies. It may permit, on the one hand, to carry out more efficient machine learning calculations by means of quantum devices, while, on the other hand, to employ machine learning techniques to better control quantum systems. Inside quantum machine learning, quantum reinforcement learning aims at developing ‘intelligent’ quantum agents that may interact with the outer world and adapt to it, with the strategy of achieving some final goal. Another paradigm inside quantum machine learning is that of quantum autoencoders, which may allow one for employing fewer resources in a quantum device via a training process. Moreover, the field of quantum biomimetics aims at establishing analogies between biological and quantum systems, to look for previously inadvertent connections that may enable useful applications. Two recent examples are the concepts of quantum artificial life, as well as of quantum memristors. In this Perspective, we give an overview of these topics, describing the related research carried out by the scientific community.
Cloud-based access to quantum computers opens up the way for the empirical implementation of quantum artificial neural networks and for the future integration of quantum computation in different devices, using the cloud to access a quantum computer. The current work experimentally implements quantum artificial neural networks on IBM’s quantum computers, accessed via cloud. Examples are provided for the XOR Boolean function representation problem and decision under risk; in the last case, quantum object-oriented programming using IBM’s Qiskit Python library is employed to implement a form of quantum neural reinforcement learning applied to a classical decision under risk problem, showing how decision can be integrated into a quantum artificial intelligence system, where an artificial agent learns how to select an optimal action when facing a classical gamble. A final reflection is provided on quantum robotics and a future where robotic systems are connected to quantum computers via cloud, using quantum neural computation to learn to optimize tasks and act accordingly.
There is no doubt artificial intelligence and machine learning have been significantly modifying the scientific and technological landscape, as well as society at large, over the last decade. Search engines, autonomous cars, language and speech translators, as well as image classifiers are only some possibilities this discipline enables and there is certainly more to come in the near future. Significant research efforts have been dedicated in the direction of quantum computing, a topic that has been receiving ever-growing attention from governmental institutions, multinational companies, and academics. A scalable quantum computer could offer unprecedented computational capabilities, enabling a revolution in information processing and transmission. Among the various approaches, quantum machine learning has emerged as an utterly successful method to better control and understand quantum systems, benefitting from machine learning algorithms, while relying on quantum processors to accelerate machine learning tasks. When coupled with biomimetics, an area of science that aims at imitating biological systems in order to design more efficient artificial devices, quantum biomimetics arises as a novel field that bridges the gap between natural selection in bio systems and engineering processes in quantum devices with one ultimate goal: The replication of biological behaviors in quantum controllable systems, what we call “bioinspired quantum technologies”. We believe that this Special Issue will shed light on unexplored areas up to now, acting as the precursor for further exciting research in the field of bioinspired quantum technologies.
The simulation of quantum matter with classical hardware plays a central role in the discovery and development of quantum many-body systems, with far-reaching implications in condensed matter physics and quantum technologies. In general, efficient and sophisticated algorithms are required to overcome the severe challenge posed by the exponential scaling of the Hilbert space of quantum systems. In contrast, hardware built with quantum bits of information are inherently capable of efficiently finding solutions of quantum many-body problems. While a universal and scalable quantum computer is still beyond the horizon, recent advances in qubit manufacturing and coherent control of synthetic quantum matter are leading to a new generation of intermediate scale quantum hardware. \n \nThe complexity underlying quantum many-body systems closely resembles the one encountered in many problems in the world of information and technology. In both contexts, the complexity stems from a large number of interacting degrees of freedom. A powerful strategy in the latter scenario is machine learning, a subfield of artificial intelligence where large amounts of data are used to extract relevant features and patterns. In particular, artificial neural networks have been demonstrated to be capable of discovering low-dimensional representations of complex objects from high-dimensional dataset, leading to the profound technological revolution we all witness in our daily life. \n \nIn this Thesis, we envision a new paradigm for scientific discovery in quantum physics. On the one hand, we have the essentially unlimited data generated with the increasing amount of highly controllable quantum hardware. On the other hand, we have a set of powerful algorithms that efficiently capture non-trivial correlations from high-dimensional data. Therefore, we fully embrace this data-driven approach to quantum mechanics, and anticipate new exciting possibilities in the field of quantum many-body physics and quantum information science. We revive a powerful stochastic neural network called a restricted Boltzmann machine, which slowly moved out of fashion after playing a central role in the machine learning revolution of the early 2010s. We introduce a neural-network representation of quantum states based on this generative model. We propose a set of algorithms to reconstruct unknown quantum states from measurement data and numerically demonstrate their potential, with important implications for current experiments. These include the reconstruction of experimentally inaccessible properties, such as entanglement, and diagnostics to determine sources of noise. Furthermore, we introduce a machine learning framework for quantum error correction, where a neural network learns the best decoding strategy directly from data. We expect that the full integration between quantum hardware and artificial intelligence will become the gold standard, and will drive the world into the era of fault-tolerant quantum computing and large-scale quantum simulations.