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

Proceedings of the AAAI Conference on Artificial Intelligence

Multi-Stage Reinforcement Learning for Robust Charging of Quantum Batteries (Student Abstract)

Beomdo Park, Hyeok Jang, Junseong Park, Minu Baek, Gihun Gil, Min Ja Jung, Woohyeon Kwon, Harin Jang, Yeojin Jang, Hoon Jeong, Taewook Heo, Sangkeum Lee

Published Mar 14, 2026• 0 citations
Publisher

Abstract

Quantum batteries have emerged as a next-generation energy storage solution, leveraging quantum phenomena such as superabsorption to overcome the limitations of conventional energy technologies. However, noise arising from interactions with the external environment degrades the charging efficiency and stability of the battery by disrupting the system's quantum coherence. To address this challenge, this study proposes a robust charging framework for a single-qubit quantum battery based on the Jaynes-Cummings (JC) model. The proposed framework combines the Proximal Policy Optimization (PPO) algorithm with a multi-stage reinforcement learning structure. The agent first learns fundamental control principles in a noise-free, ideal environment and subsequently performs robust learning in progressively noisier and more complex settings. Simulation results demonstrate that the trained agent navigates a stable charging trajectory on the Bloch sphere, thereby achieving high ergotropy even in the presence of noise. These findings suggest that multi-stage reinforcement learning is an effective solution for control problems in noisy quantum systems and provides a theoretical foundation for designing charging protocols for multi-qubit systems.

DOI
10.1609/aaai.v40i48.42264

Related papers

Optimizing the Charging of Open Quantum Batteries using Long Short-Term Memory-Driven Reinforcement Learning

Shadab Zakavati, S. Salimi, Behrouz Arash

Controlling the charging process of a quantum battery involves strategies to efficiently transfer, store, and retain energy, while mitigating decoherence, energy dissipation, and inefficiencies caused by surrounding interactions. We develop a model to study the charging process of a quantum battery in an open quantum setting, where the battery interacts with a charger and a structured reservoir. To overcome the limitations of static charging protocols, a reinforcement learning (RL) charging strategy is proposed, which utilizes the deep deterministic policy gradient algorithm alongside long short-term memory (LSTM) networks. The LSTM networks enable the RL model to capture temporal correlations driven by non-Markovian dynamics, facilitating a continuous, adaptive charging strategy. The RL protocols consistently outperform conventional fixed heuristic strategies by real-time controlling the driving field amplitude and coupling parameters. By penalizing battery-to-charger backflow in the reward function, the RL-optimized charging strategy promotes efficient unidirectional energy transfer from charger to battery, achieving higher and more stable extractable work. The proposed RL controller would provide a framework for designing efficient charging schemes in broader configurations and multi-cell quantum batteries.

arXiv:2504.19840•Apr 28, 2025•0 citations

Adaptive Reinforcement Learning for Robust Open Quantum System Control: A Multi-Task Framework with Temporal Optimization

Haftu W. Fentaw, Steve Campbell, Simon Caton

We present a Multi-task Soft Actor-Critic (SAC) Reinforcement Learning framework designed for open-system quantum control across diverse Hamiltonians, which learns optimal pulse sequences while simultaneously discovering problem-specific evolution time T and number of control pulse segments N. Experimental results across 51 Hamiltonian variations demonstrate that the multi-task SAC model is able to generate control pulses that can drive a system, under environment noise, from its initial state to its target state with high fidelities, establishing essential foundations for universal quantum control applicable to realistic noisy quantum devices. Through progressive expansion of the training Hamiltonian set, we investigate if a single multi-task model trained using a given number of sample Hamiltonians can successfully accomplish state-transfer tasks for Hamiltonians drawn from the same Hamiltonian space but not encountered during training. In addition, our Robustness Infidelity Measure (RIM) analysis reveals that SAC trained policies exhibit superior robustness to pulse amplitude perturbations and decoherence rate variations compared to GRAPE-optimized controls.

quant-pharXiv:2605.26925•May 27, 2026•0 citations

Advanced Quantum Control With Ensemble Reinforcement Learning: A Case Study on the XY Spin Chain

Farshad Rahimi Ghashghaei, Nebrase Elmrabit, Ayyaz-Ul-Haq Qureshi +2 more

This research presents an ensemble Reinforcement Learning (RL) approach that combines Deep Q-Network (DQN) and Proximal Policy Optimization (PPO) algorithms to tackle quantum control problems. This research aims to use the complementary strengths of DQN and PPO algorithms to develop robust and adaptive control policies for noisy and uncertain quantum systems. We comprehensively analyse the proposed ensemble learning, including algorithmic details, implementation specifics, and experimental results. Through extensive experimentation and evaluation, we demonstrate the effectiveness of the ensemble approach in learning control strategies for manipulating quantum systems towards a random target state. The results highlight the potential of ensemble RL techniques in addressing the challenges of quantum control tasks, such as system noise and dynamics. By integrating multiple RL agents within an ensemble framework, We aim to advance current developments in quantum control and create a new path for the development of adaptive control systems for quantum systems. The performance of the ensemble model is assessed against Gradient Ascent Pulse Engineering (GRAPE) and robust Model Predictive Control (MPC) to demonstrate its efficiency in highly challenging and noisy environments.

IEEE Access•Jan 1, 2025•1 citation

Reinforcement Learning for Charging Optimization of Inhomogeneous Dicke Quantum Batteries

Xiaobin Song, Siyuan Bai, Da-Wei Wang +4 more

Charging optimization is a key challenge to the implementation of quantum batteries, particularly under inhomogeneity and partial observability. This paper employs reinforcement learning to optimize piecewise-constant charging policies for an inhomogeneous Dicke battery. We systematically compare policies across four observability regimes, from full-state access to experimentally accessible observables (energies of individual two-level systems (TLSs), first-order averages, and second-order correlations). Simulation results demonstrate that full observability yields near-optimal ergotropy with low variability, while under partial observability, access to only single-TLS energies or energies plus first-order averages lags behind the fully observed baseline. However, augmenting partial observations with second-order correlations recovers most of the gap, reaching 94%-98% of the full-state baseline. The learned schedules are nonmyopic, trading temporary plateaus or declines for superior terminal outcomes. These findings highlight a practical route to effective fast-charging protocols under realistic information constraints.

quant-pharXiv:2511.12176•Cybernetics and Intelligence•Nov 18, 2025•0 citations

Maximum-precision charging of multi-qubit quantum batteries

Davide Rinaldi, Radim Filip, Dario Gerace +1 more

Precision, robustness, and efficiency are central requirements for quantum technologies. We show that genuine quantum features combined with non-Gaussianity enable the simultaneous optimization of these properties in a quantum battery-charging process. Using a generalized Jaynes-Cummings interaction as a paradigmatic light-matter interaction model, we apply the Full Counting Statistics to characterize stochastic energy exchanges between a stack of qubits and a single-mode bosonic field. We demonstrate that a sequential charging protocol driven by a non-Gaussian quantum field yields high performance in charging precision, which remains maximal even under suboptimal operating conditions. Our results establish the use of non-Gaussian quantum-states in battery charging as a robust route to a quantum precision advantage over protocols based on Gaussian states, achieved through the suppression of detrimental quantum fluctuations.

quant-pharXiv:2601.12183•Physical Review A•Jan 20, 2026•1 citation

Criticality-Enhanced Charging of Quantum Battery via Optimized Quantum Amplification

Jiawei Zhang, Teng Liu, M. Li +3 more

Abstract Achieving rapid charging in quantum batteries requires the efficient harnessing of quantum resources, which is a task with great challenge experimentally. Here we propose and experimentally realize a quantum battery charging protocol that enables fast injection of substantial energy using a trapped-ion platform governed by the quantum Rabi model (QRM). The core of the protocol lies in the employment of the squeezing effect and the QRM criticality. Reinforcement learning further improves the charging performance by optimizing the preparation of the squeezed and anti-squeezed states and accelerating energy inflow near criticality. Experimental implementation with a single ${}^{171}\mathrm{Yb}^+$ ion demonstrates the feasibility of the protocol, where the vibrational mode (i.e., quantized phonons) acts as the quantum battery. We witness that both the maximally extractable work (i.e., ergotropy) and the charging power exhibit criticality-induced enhancement. This work provides a practical pathway toward criticality-enhanced quantum batteries, underscoring the potential of squeezing-mediated amplification and displacement-based energy storage in realistic quantum systems.

•Apr 22, 2026•0 citations
Quantum batteries have emerged as a next-generation energy storage solution, leveraging quantum phenomena such as superabsorption to overcome the limitations of conventional energy technologies. However, noise arising from interactions with the external environment degrades the charging efficiency and stability of the battery by disrupting the system's quantum coherence. To address this challenge, this study proposes a robust charging framework for a single-qubit quantum battery based on the Jaynes-Cummings (JC) model. The proposed framework combines the Proximal Policy Optimization (PPO) algorithm with a multi-stage reinforcement learning structure. The agent first learns fundamental control principles in a noise-free, ideal environment and subsequently performs robust learning in progressively noisier and more complex settings. Simulation results demonstrate that the trained agent navigates a stable charging trajectory on the Bloch sphere, thereby achieving high ergotropy even in the presence of noise. These findings suggest that multi-stage reinforcement learning is an effective solution for control problems in noisy quantum systems and provides a theoretical foundation for designing charging protocols for multi-qubit systems.
Controlling the charging process of a quantum battery involves strategies to efficiently transfer, store, and retain energy, while mitigating decoherence, energy dissipation, and inefficiencies caused by surrounding interactions. We develop a model to study the charging process of a quantum battery in an open quantum setting, where the battery interacts with a charger and a structured reservoir. To overcome the limitations of static charging protocols, a reinforcement learning (RL) charging strategy is proposed, which utilizes the deep deterministic policy gradient algorithm alongside long short-term memory (LSTM) networks. The LSTM networks enable the RL model to capture temporal correlations driven by non-Markovian dynamics, facilitating a continuous, adaptive charging strategy. The RL protocols consistently outperform conventional fixed heuristic strategies by real-time controlling the driving field amplitude and coupling parameters. By penalizing battery-to-charger backflow in the reward function, the RL-optimized charging strategy promotes efficient unidirectional energy transfer from charger to battery, achieving higher and more stable extractable work. The proposed RL controller would provide a framework for designing efficient charging schemes in broader configurations and multi-cell quantum batteries.
We present a Multi-task Soft Actor-Critic (SAC) Reinforcement Learning framework designed for open-system quantum control across diverse Hamiltonians, which learns optimal pulse sequences while simultaneously discovering problem-specific evolution time T and number of control pulse segments N. Experimental results across 51 Hamiltonian variations demonstrate that the multi-task SAC model is able to generate control pulses that can drive a system, under environment noise, from its initial state to its target state with high fidelities, establishing essential foundations for universal quantum control applicable to realistic noisy quantum devices. Through progressive expansion of the training Hamiltonian set, we investigate if a single multi-task model trained using a given number of sample Hamiltonians can successfully accomplish state-transfer tasks for Hamiltonians drawn from the same Hamiltonian space but not encountered during training. In addition, our Robustness Infidelity Measure (RIM) analysis reveals that SAC trained policies exhibit superior robustness to pulse amplitude perturbations and decoherence rate variations compared to GRAPE-optimized controls.
This research presents an ensemble Reinforcement Learning (RL) approach that combines Deep Q-Network (DQN) and Proximal Policy Optimization (PPO) algorithms to tackle quantum control problems. This research aims to use the complementary strengths of DQN and PPO algorithms to develop robust and adaptive control policies for noisy and uncertain quantum systems. We comprehensively analyse the proposed ensemble learning, including algorithmic details, implementation specifics, and experimental results. Through extensive experimentation and evaluation, we demonstrate the effectiveness of the ensemble approach in learning control strategies for manipulating quantum systems towards a random target state. The results highlight the potential of ensemble RL techniques in addressing the challenges of quantum control tasks, such as system noise and dynamics. By integrating multiple RL agents within an ensemble framework, We aim to advance current developments in quantum control and create a new path for the development of adaptive control systems for quantum systems. The performance of the ensemble model is assessed against Gradient Ascent Pulse Engineering (GRAPE) and robust Model Predictive Control (MPC) to demonstrate its efficiency in highly challenging and noisy environments.
Charging optimization is a key challenge to the implementation of quantum batteries, particularly under inhomogeneity and partial observability. This paper employs reinforcement learning to optimize piecewise-constant charging policies for an inhomogeneous Dicke battery. We systematically compare policies across four observability regimes, from full-state access to experimentally accessible observables (energies of individual two-level systems (TLSs), first-order averages, and second-order correlations). Simulation results demonstrate that full observability yields near-optimal ergotropy with low variability, while under partial observability, access to only single-TLS energies or energies plus first-order averages lags behind the fully observed baseline. However, augmenting partial observations with second-order correlations recovers most of the gap, reaching 94%-98% of the full-state baseline. The learned schedules are nonmyopic, trading temporary plateaus or declines for superior terminal outcomes. These findings highlight a practical route to effective fast-charging protocols under realistic information constraints.
Precision, robustness, and efficiency are central requirements for quantum technologies. We show that genuine quantum features combined with non-Gaussianity enable the simultaneous optimization of these properties in a quantum battery-charging process. Using a generalized Jaynes-Cummings interaction as a paradigmatic light-matter interaction model, we apply the Full Counting Statistics to characterize stochastic energy exchanges between a stack of qubits and a single-mode bosonic field. We demonstrate that a sequential charging protocol driven by a non-Gaussian quantum field yields high performance in charging precision, which remains maximal even under suboptimal operating conditions. Our results establish the use of non-Gaussian quantum-states in battery charging as a robust route to a quantum precision advantage over protocols based on Gaussian states, achieved through the suppression of detrimental quantum fluctuations.
Abstract Achieving rapid charging in quantum batteries requires the efficient harnessing of quantum resources, which is a task with great challenge experimentally. Here we propose and experimentally realize a quantum battery charging protocol that enables fast injection of substantial energy using a trapped-ion platform governed by the quantum Rabi model (QRM). The core of the protocol lies in the employment of the squeezing effect and the QRM criticality. Reinforcement learning further improves the charging performance by optimizing the preparation of the squeezed and anti-squeezed states and accelerating energy inflow near criticality. Experimental implementation with a single 171Yb+{}^{171}\mathrm{Yb}^+171Yb+ ion demonstrates the feasibility of the protocol, where the vibrational mode (i.e., quantized phonons) acts as the quantum battery. We witness that both the maximally extractable work (i.e., ergotropy) and the charging power exhibit criticality-induced enhancement. This work provides a practical pathway toward criticality-enhanced quantum batteries, underscoring the potential of squeezing-mediated amplification and displacement-based energy storage in realistic quantum systems.