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

Uhlmann-Geometric Quantum Control of Attosecond Molecular Dynamics: A Decoherence-Resistant Framework via Hybrid Reinforcement Learning

mohamed kamal arafa el-rakhawi

Published Nov 7, 2026• 0 citations
Publisher

Abstract

Controlling quantum molecular dynamics remains fundamentally challenged by environmental decoherence. We introduce Uhlmann-Geometric Quantum Control (UGQC), a novel framework integrating geometry-aware robustness with hybrid AI-driven pulse optimization. The key innovation is the Uhlmann Geometric Robustness Metric (UGRM), which measures fidelity between noisy and ideal trajectories evolving under identical control fields, implicitly stabilizing the Uhlmann phase to provide geometric protection against decoherence. Our hybrid approach combines GRAPE-generated physically informed initial fields with Soft Actor-Critic reinforcement learning fine-tuning, effectively navigating non-convex control landscapes. Validated on a four-level molecular system with strong dissipation rates simulating condensed-phase conditions, UGQC achieves 89% target state fidelity versus 62% for standard optimal control, while maintaining UGRM above 0.91. Under additional stochastic perturbations, our framework exhibits only 5% performance degradation compared to 27% for conventional methods. This work establishes a mathematically rigorous, computationally feasible pathway toward decoherence-resistant attosecond chemistry, with clear experimental implementation using contemporary pulse shaping technologies. The complete reproducible code is archived under DOI: 10.5281/zenodo.24033893.

DOI
10.5281/zenodo.24033892

Related papers

Topological Quantum Control of Attosecond Molecular Dynamics: A Proof-of-Principle Framework on Reduced-Dimensional Models

mohamed kamal arafa el-rakhawi

The El-Rakhawi Framework for Topological Quantum Control (DOI: 10.5281/zenodo.22907946) presents a proof-of-principle methodology integrating Uhlmann-based robustness metrics with Reinforcement Learning (PPO) to control open quantum systems governed by the Lindblad master equation. By embedding an Uhlmann Robustness Metric (URM) into the optimization reward function, the framework guides molecular dynamics along trajectories resilient to severe environmental decoherence. Simulations on a 3-level Lambda system demonstrate a target state yield improvement from ~62% (standard OCT) to ~87%. An ablation study confirms this gain is directly attributable to the URM constraint. The work maintains strict scientific honesty, acknowledging this is a reduced-dimensionality model with noted computational considerations for larger systems. Protected under sovereign intellectual property with explicit AI training restrictions and mandatory academic attribution. All rights exclusively reserved to Dr. Mohamed Kamal Arafa El-Rakhawi, establishing a reproducible blueprint for decoherence-resistant quantum control.

•Oct 15, 2026•0 citations

Topological Quantum Control of Attosecond Molecular Dynamics: A Proof-of-Principle Framework on Reduced-Dimensional Models

m el-rakhawi

This paper introduces Topological Quantum Control (TQC), a framework integrating Uhlmann fidelity-based robustness metrics with AI-driven pulse optimization for attosecond molecular dynamics. Using the Lindblad master equation to model decoherence, we employ Proximal Policy Optimization (PPO) to navigate non-convex control landscapes. Validated on a three-level Lambda system under extreme noise stress-tests, TQC improves target yield from 62% to 87% compared to standard optimal control. The approach demonstrates enhanced resilience against environmental decoherence, offering a reproducible, open-source proof-of-principle for robust quantum control in complex, open systems.

•Sep 19, 2026•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

Topological Quantum Control of Attosecond Molecular Dynamics: A Proof-of-Principle Framework on Reduced-Dimensional Models

m el-rakhawi

ABSTRACT Femtochemistry enabled observation of atomic motion, yet controlling chemical reactions with high selectivity remains challenging due to environmental decoherence and the probabilistic nature of quantum dynamics. This paper proposes Topological Quantum Control (TQC), a theoretical framework that integrates topological invariants of open quantum systems with AI-driven pulse optimization. We formulate the control problem using the Lindblad master equation and introduce a Topological Robustness Index (TRI) based on the Uhlmann fidelity to guide optimization toward trajectories that are resilient to decoherence. We derive the adjoint-state gradient for the topological objective functional and implement a reinforcement learning agent for pulse shaping. As a proof of principle, we validate the framework on a three-level Lambda-type model system. Simulations suggest that the TQC framework can improve target product yield from approximately 62 percent under standard optimal control to approximately 87 percent under the proposed topological constraint, while maintaining high TRI values throughout the evolution. These results are obtained on a reduced-dimensionality model and serve as a conceptual demonstration; extension to realistic molecular systems remains a subject of future work. Keywords: Attosecond Physics, Topological Quantum Control, Lindblad Master Equation, Uhlmann Phase, Quantum Optimal Control, Reinforcement Learning.

•Aug 3, 2026•0 citations

Uncovering Latent Structures in Robust Pulse Sequences: A Model-Based Reinforcement Learning Approach for Adaptable Quantum Control

Tobias Kiermeyer, Thomas Heydenreich, Léo Van Damme +3 more

Real-time adaptive control of quantum systems requires rapid generation of robust, high-fidelity pulses across a continuous range of operating conditions. Standard optimization algorithms such as gradient-ascent pulse engineering (GRAPE) solve each instance independently, discarding information between runs and requiring costly reinitialization when parameters change. We present an approach to robust optimal quantum control based on model-based reinforcement learning, in which a single neural network -- embedding the Hamiltonian directly into the training pipeline -- generates robust gates across an entire family of gate configurations, without pre-computed training data. Demonstrated on a single-spin (two-level) system, the trained networks produce pulses for arbitrary rotation angles over a range of pulse durations, detunings, and field inhomogeneities in milliseconds, at fidelities comparable to multi-seed GRAPE. The framework is inherently adaptable: any parameter entering the Hamiltonian can serve as a network input, extending the approach to different systems and control settings. Beyond speed, the network reveals structure in the control landscape: it discovers the same structured phase profiles that appear in GRAPE solutions -- made identifiable through fidelity-invariant symmetry transformations -- but more consistently than independent optimization. This consistency enables smooth interpolation across the entire trained parameter space.

quant-pharXiv:2606.24507•Jun 24, 2026•2 citations

Dynamically Corrected Quantum Control: A Geometrical Framework

Junkai Zeng

Implementing high-fidelity quantum control and suppressing the unwanted environmental noise has been one of the essential challenges in developing quantum information technologies. In the past, driving pulse sequences based on Dirac delta functions or square wave functions, such as Hahn spin echo or CPMG, have been developed to dynamically correcting the noise effects. However, implementing these ideal pulses with high fidelity is a challenging task in real experiments. In this thesis, we provide a new and simple method to explore the entire solution space of driving pulse shapes that suppress environmental noise in the evolution of the system. In this method, any single-qubit phase gate that is first-order robust against quasi-static transversal noise corresponds to a closed curve on a two-dimensional plane, and more general first-order robust single-qubit gates correspond to closed three-dimensional space curves. Second-order robust gates correspond to closed curves having the property that their projection onto any two-dimensional planes shall enclose a zero net area. The driving pulse shapes that implement the gates can be determined by the curvature, torsion, and the length of the curve. By utilizing the framework it is possible to obtain globally optimal solutions in pulse shaping in respect of experimental constraints by mapping them into geometrical optimization problems. One such problem we solved is to prove that the fastest possible single-qubit phase gates that are second-order noise-resistant shall be implemented using sign-flipping square functions. Since square waves are not experimentally feasible, we provide a method to smooth these pulses with minimal loss in gate speed while maintaining the robustness, based on the geometrical framework. This framework can also be useful in diagnosing the noise-cancellation properties of pulse shapes generated from numerical methods such as GRAPE. We show that this method for pulse shaping can significantly improve the fidelity of single-qubit gates through numerical simulation.

•Oct 22, 2019•0 citations
Controlling quantum molecular dynamics remains fundamentally challenged by environmental decoherence. We introduce Uhlmann-Geometric Quantum Control (UGQC), a novel framework integrating geometry-aware robustness with hybrid AI-driven pulse optimization. The key innovation is the Uhlmann Geometric Robustness Metric (UGRM), which measures fidelity between noisy and ideal trajectories evolving under identical control fields, implicitly stabilizing the Uhlmann phase to provide geometric protection against decoherence. Our hybrid approach combines GRAPE-generated physically informed initial fields with Soft Actor-Critic reinforcement learning fine-tuning, effectively navigating non-convex control landscapes. Validated on a four-level molecular system with strong dissipation rates simulating condensed-phase conditions, UGQC achieves 89% target state fidelity versus 62% for standard optimal control, while maintaining UGRM above 0.91. Under additional stochastic perturbations, our framework exhibits only 5% performance degradation compared to 27% for conventional methods. This work establishes a mathematically rigorous, computationally feasible pathway toward decoherence-resistant attosecond chemistry, with clear experimental implementation using contemporary pulse shaping technologies. The complete reproducible code is archived under DOI: 10.5281/zenodo.24033893.
The El-Rakhawi Framework for Topological Quantum Control (DOI: 10.5281/zenodo.22907946) presents a proof-of-principle methodology integrating Uhlmann-based robustness metrics with Reinforcement Learning (PPO) to control open quantum systems governed by the Lindblad master equation. By embedding an Uhlmann Robustness Metric (URM) into the optimization reward function, the framework guides molecular dynamics along trajectories resilient to severe environmental decoherence. Simulations on a 3-level Lambda system demonstrate a target state yield improvement from ~62% (standard OCT) to ~87%. An ablation study confirms this gain is directly attributable to the URM constraint. The work maintains strict scientific honesty, acknowledging this is a reduced-dimensionality model with noted computational considerations for larger systems. Protected under sovereign intellectual property with explicit AI training restrictions and mandatory academic attribution. All rights exclusively reserved to Dr. Mohamed Kamal Arafa El-Rakhawi, establishing a reproducible blueprint for decoherence-resistant quantum control.
This paper introduces Topological Quantum Control (TQC), a framework integrating Uhlmann fidelity-based robustness metrics with AI-driven pulse optimization for attosecond molecular dynamics. Using the Lindblad master equation to model decoherence, we employ Proximal Policy Optimization (PPO) to navigate non-convex control landscapes. Validated on a three-level Lambda system under extreme noise stress-tests, TQC improves target yield from 62% to 87% compared to standard optimal control. The approach demonstrates enhanced resilience against environmental decoherence, offering a reproducible, open-source proof-of-principle for robust quantum control in complex, open systems.
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.
ABSTRACT Femtochemistry enabled observation of atomic motion, yet controlling chemical reactions with high selectivity remains challenging due to environmental decoherence and the probabilistic nature of quantum dynamics. This paper proposes Topological Quantum Control (TQC), a theoretical framework that integrates topological invariants of open quantum systems with AI-driven pulse optimization. We formulate the control problem using the Lindblad master equation and introduce a Topological Robustness Index (TRI) based on the Uhlmann fidelity to guide optimization toward trajectories that are resilient to decoherence. We derive the adjoint-state gradient for the topological objective functional and implement a reinforcement learning agent for pulse shaping. As a proof of principle, we validate the framework on a three-level Lambda-type model system. Simulations suggest that the TQC framework can improve target product yield from approximately 62 percent under standard optimal control to approximately 87 percent under the proposed topological constraint, while maintaining high TRI values throughout the evolution. These results are obtained on a reduced-dimensionality model and serve as a conceptual demonstration; extension to realistic molecular systems remains a subject of future work. Keywords: Attosecond Physics, Topological Quantum Control, Lindblad Master Equation, Uhlmann Phase, Quantum Optimal Control, Reinforcement Learning.
Real-time adaptive control of quantum systems requires rapid generation of robust, high-fidelity pulses across a continuous range of operating conditions. Standard optimization algorithms such as gradient-ascent pulse engineering (GRAPE) solve each instance independently, discarding information between runs and requiring costly reinitialization when parameters change. We present an approach to robust optimal quantum control based on model-based reinforcement learning, in which a single neural network -- embedding the Hamiltonian directly into the training pipeline -- generates robust gates across an entire family of gate configurations, without pre-computed training data. Demonstrated on a single-spin (two-level) system, the trained networks produce pulses for arbitrary rotation angles over a range of pulse durations, detunings, and field inhomogeneities in milliseconds, at fidelities comparable to multi-seed GRAPE. The framework is inherently adaptable: any parameter entering the Hamiltonian can serve as a network input, extending the approach to different systems and control settings. Beyond speed, the network reveals structure in the control landscape: it discovers the same structured phase profiles that appear in GRAPE solutions -- made identifiable through fidelity-invariant symmetry transformations -- but more consistently than independent optimization. This consistency enables smooth interpolation across the entire trained parameter space.
Implementing high-fidelity quantum control and suppressing the unwanted environmental noise has been one of the essential challenges in developing quantum information technologies. In the past, driving pulse sequences based on Dirac delta functions or square wave functions, such as Hahn spin echo or CPMG, have been developed to dynamically correcting the noise effects. However, implementing these ideal pulses with high fidelity is a challenging task in real experiments. In this thesis, we provide a new and simple method to explore the entire solution space of driving pulse shapes that suppress environmental noise in the evolution of the system. In this method, any single-qubit phase gate that is first-order robust against quasi-static transversal noise corresponds to a closed curve on a two-dimensional plane, and more general first-order robust single-qubit gates correspond to closed three-dimensional space curves. Second-order robust gates correspond to closed curves having the property that their projection onto any two-dimensional planes shall enclose a zero net area. The driving pulse shapes that implement the gates can be determined by the curvature, torsion, and the length of the curve. By utilizing the framework it is possible to obtain globally optimal solutions in pulse shaping in respect of experimental constraints by mapping them into geometrical optimization problems. One such problem we solved is to prove that the fastest possible single-qubit phase gates that are second-order noise-resistant shall be implemented using sign-flipping square functions. Since square waves are not experimentally feasible, we provide a method to smooth these pulses with minimal loss in gate speed while maintaining the robustness, based on the geometrical framework. This framework can also be useful in diagnosing the noise-cancellation properties of pulse shapes generated from numerical methods such as GRAPE. We show that this method for pulse shaping can significantly improve the fidelity of single-qubit gates through numerical simulation.