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

Quantum machine learning

Riu Vicente, Jordi

Published Apr 14, 2019• 0 citations
OpenAlex

Abstract

We use reinforcement learning techniques to optimize the Quantum Approximate Optimization Algorithm when applied to the MaxCut problem. We explore Q-learning based techniques both for continuous and discrete action environments with regular and irregular graphs.

Related papers

Quantum Machine Learning

Quantum Machine Learningby Jordi RIU This project applies reinforcement learning techniques to optimize a quantum algorithm.The upcoming quantum computers are capable of optimization tasks in an exponentially large parameter space.Having access to such a large optimization space potentially allows an advantage of Quantum methods over classical algorithms, but at the price of problems in the optimization process -just to name a few, the presence of local minima and the low convergence speeds-.This project explores the application of reinforcement learning techniques to control and optimize the operation of a quantum computer solving a hard combinatorial problem, namely the MaxCut problem.Our methods use a collection of quantum observables to feed the state of the agent, inspired on a similar classical approach to the problem applied to the antiferromagnetic Ising model lattice.Our results show how an agent using these observables can optimally control the operation of a quantum device to obtain quasi-optimal solutions when in combination with the Quantum Approximate Optimization Algorithm (QAOA).v

Between science and economics•Apr 1, 2020•9 citations

Reinforcement Learning in Different Phases of Quantum Control

Marin Bukov, Alexandre G. R. Day, Dries Sels +3 more

New experiments show that reinforcement learning algorithms, a cutting-edge technique for machine learning, can quickly and accurately learn to prepare a desired quantum state despite no knowledge of quantum mechanics.

arXiv:1705.00565•Physical Review X•Sep 27, 2018•333 citations

Quantum Observables for continuous control of the Quantum Approximate Optimization Algorithm via Reinforcement Learning

Artur García-Sáez, Riu, Jordi

We present a classical control mechanism for Quantum devices using Reinforcement Learning. Our strategy is applied to the Quantum Approximate Optimization Algorithm (QAOA) in order to optimize an objective function that encodes a solution to a hard combinatorial problem. This method provides optimal control of the Quantum device following a reformulation of QAOA as an environment where an autonomous classical agent interacts and performs actions to achieve higher rewards. This formulation allows a hybrid classical-Quantum device to train itself from previous executions using a continuous formulation of deep Q-learning to control the continuous degrees of freedom of QAOA. Our approach makes a selective use of Quantum measurements to complete the observations of the Quantum state available to the agent. We run tests of this approach on MAXCUT instances of size up to N = 21 obtaining optimal results. We show how this formulation can be used to transfer the knowledge from shorter training episodes to reach a regime of longer executions where QAOA delivers higher results.

arXiv:1911.09682•Nov 21, 2019•13 citations

An improved adaptive quantum genetic algorithm as classical optimizer for the quantum approximate optimization algorithm on MaxCut problem

Yu Zhang, Jing Zhao, Yanguo Jia +1 more

Quantum Information Processing•Jul 12, 2025•3 citations

Quantum Machine Learning for Optimization Problems Leveraging Quantum Computing to Speed Up Large-Scale Optimization in AI Models

Ram Awtar, Rohitaswa Pandey

International Conference on Green Technologies and Sustainable Solutions (GTSS)•Feb 1, 2026•0 citations

Graph neural network initialisation of quantum approximate optimisation

Nishant Jain, Brian Coyle, Elham Kashefi +1 more

Approximate combinatorial optimisation has emerged as one of the most promising application areas for quantum computers, particularly those in the near term. In this work, we focus on the quantum approximate optimisation algorithm (QAOA) for solving the MaxCut problem. Specifically, we address two problems in the QAOA, how to initialise the algorithm, and how to subsequently train the parameters to find an optimal solution. For the former, we propose graph neural networks (GNNs) as a warm-starting technique for QAOA. We demonstrate that merging GNNs with QAOA can outperform both approaches individually. Furthermore, we demonstrate how graph neural networks enables warm-start generalisation across not only graph instances, but also to increasing graph sizes, a feature not straightforwardly available to other warm-starting methods. For training the QAOA, we test several optimisers for the MaxCut problem up to 16 qubits and benchmark against vanilla gradient descent. These include quantum aware/agnostic and machine learning based/neural optimisers. Examples of the latter include reinforcement and meta-learning. With the incorporation of these initialisation and optimisation toolkits, we demonstrate how the optimisation problems can be solved using QAOA in an end-to-end differentiable pipeline.

arXiv:2111.03016•Quantum•Nov 17, 2022•49 citations
Quantum Machine Learningby Jordi RIU This project applies reinforcement learning techniques to optimize a quantum algorithm.The upcoming quantum computers are capable of optimization tasks in an exponentially large parameter space.Having access to such a large optimization space potentially allows an advantage of Quantum methods over classical algorithms, but at the price of problems in the optimization process -just to name a few, the presence of local minima and the low convergence speeds-.This project explores the application of reinforcement learning techniques to control and optimize the operation of a quantum computer solving a hard combinatorial problem, namely the MaxCut problem.Our methods use a collection of quantum observables to feed the state of the agent, inspired on a similar classical approach to the problem applied to the antiferromagnetic Ising model lattice.Our results show how an agent using these observables can optimally control the operation of a quantum device to obtain quasi-optimal solutions when in combination with the Quantum Approximate Optimization Algorithm (QAOA).v
We present a classical control mechanism for Quantum devices using Reinforcement Learning. Our strategy is applied to the Quantum Approximate Optimization Algorithm (QAOA) in order to optimize an objective function that encodes a solution to a hard combinatorial problem. This method provides optimal control of the Quantum device following a reformulation of QAOA as an environment where an autonomous classical agent interacts and performs actions to achieve higher rewards. This formulation allows a hybrid classical-Quantum device to train itself from previous executions using a continuous formulation of deep Q-learning to control the continuous degrees of freedom of QAOA. Our approach makes a selective use of Quantum measurements to complete the observations of the Quantum state available to the agent. We run tests of this approach on MAXCUT instances of size up to N = 21 obtaining optimal results. We show how this formulation can be used to transfer the knowledge from shorter training episodes to reach a regime of longer executions where QAOA delivers higher results.
Approximate combinatorial optimisation has emerged as one of the most promising application areas for quantum computers, particularly those in the near term. In this work, we focus on the quantum approximate optimisation algorithm (QAOA) for solving the MaxCut problem. Specifically, we address two problems in the QAOA, how to initialise the algorithm, and how to subsequently train the parameters to find an optimal solution. For the former, we propose graph neural networks (GNNs) as a warm-starting technique for QAOA. We demonstrate that merging GNNs with QAOA can outperform both approaches individually. Furthermore, we demonstrate how graph neural networks enables warm-start generalisation across not only graph instances, but also to increasing graph sizes, a feature not straightforwardly available to other warm-starting methods. For training the QAOA, we test several optimisers for the MaxCut problem up to 16 qubits and benchmark against vanilla gradient descent. These include quantum aware/agnostic and machine learning based/neural optimisers. Examples of the latter include reinforcement and meta-learning. With the incorporation of these initialisation and optimisation toolkits, we demonstrate how the optimisation problems can be solved using QAOA in an end-to-end differentiable pipeline.