arXiv subjects

Michael Schilling

Publications and source records attributed to Michael Schilling.

6 recordsLinked to original sources

arXiv2025

Gradients, parallelism, and variance of quantum estimates

Computation of observables and their gradients on near-term quantum hardware is a central aspect of any quantum algorithm. In this work, we first review standard approaches to the estimation of observables with and without quantum amplitude estimation for both cost functions and gradients, discuss sampling problems, and analyze variance propagation on quantum circuits with and without Linear Combination of Unitaries (LCU). Afterwards, we systematically analyze the standard approaches to gradient computation with LCU circuits. Finally, we develop a LCU gradient framework for the most general gradients based on n-qubit gates and for time-dependent quantum control gradient, analyze the convergence behaviour of the circuit estimators, and provide detailed circuit representations of both for near-term and fault-tolerant hardware.

quant-ph↗

arXiv2025

Real-time adaptive quantum error correction by model-free multi-agent learning

Quantum error correction (QEC) is essential for scalable quantum computing, yet existing approaches rely on static assumptions about noise that break down in realistic hardware, where error channels drift over time. We introduce a unified framework that separates QEC into two learning timescales: offline code discovery and online adaptation. Offline, Multi-Agent Reinforcement Learning (MARL) autonomously discovers complete QEC cycles as explicit quantum circuits, with separate agents responsible for encoding, syndrome extraction, and error recovery, and without prescribing a code family or circuit ansatz. Online, a lightweight adaptive layer, termed Bandit Retraining for Adaptive Variational Error Correction (BRAVE), continuously retunes a low-dimensional variational parameterization without retraining the full MARL stack. This yields a "discover once, adapt continuously" strategy that combines the flexibility of learned codes with real-time adaptation to non-stationary noise. At sufficiently high sampling rates relative to the noise drift, our method reduces logical infidelity by roughly 18-fold for qubit codes and 3-fold for qutrit codes compared to static error correction, while substantially extending robustness to noise fluctuations. These results establish a paradigm in which QEC is no longer static but is dynamically optimized for realistic quantum hardware.

quant-ph↗

arXiv2024

Exponentiation of Parametric Hamiltonians via Unitary interpolation

The effort to generate matrix exponentials and associated differentials, required to determine the time evolution of quantum systems, frequently constrains the evaluation of problems in quantum control theory, variational circuit compilation, or Monte-Carlo sampling. We introduce two ideas for the time-efficient approximation of matrix exponentials of linear multi-parametric Hamiltonians. We modify the Suzuki-Trotter product formula from an approximation to an interpolation schemes to improve both accuracy and computational time. This allows us to achieve high fidelities within a single interpolation step, which can be computed directly from cached matrices. We furthermore define the interpolation on a grid of system parameters, and show that the infidelity of the interpolation converges with $4^\mathrm{th}$ order in the number of interpolation bins.

quant-ph↗

arXiv2024

Analytical solutions for optimal photon absorption into inhomogeneous spin memories

We investigate for optimal photon absorption a quantum electrodynamical model of an inhomogeneously-broadened spin ensemble coupled to a single-mode cavity. Solutions to this problem under experimental assumptions are developed in the Schr\"odinger picture without using perturbation theory concerning the cavity-spin interactions. Furthermore, we exploit the possibility of modulating the frequency and coupling rate of the resonator. We consider a one-photon input pulse and show some optimal scenarios, where exact formulas and numerical results are obtained for the absorption probabilities and the optimal pulse shapes. In particular, if the external loss dominates the internal loss of the cavity, we find the optimal cooperativity for different parameters and identify cases where absorption with a success probability larger than $99\%$ is achieved.

quant-ph↗

arXiv2023

Hybrid discrete-continuous compilation of trapped-ion quantum circuits with deep reinforcement learning

Shortening quantum circuits is crucial to reducing the destructive effect of environmental decoherence and enabling useful algorithms. Here, we demonstrate an improvement in such compilation tasks via a combination of using hybrid discrete-continuous optimization across a continuous gate set, and architecture-tailored implementation. The continuous parameters are discovered with a gradient-based optimization algorithm, while in tandem the optimal gate orderings are learned via a deep reinforcement learning algorithm, based on projective simulation. To test this approach, we introduce a framework to simulate collective gates in trapped-ion systems efficiently on a classical device. The algorithm proves able to significantly reduce the size of relevant quantum circuits for trapped-ion computing. Furthermore, we show that our framework can also be applied to an experimental setup whose goal is to reproduce an unknown unitary process.

quant-ph↗

arXiv2017

Studying the Impact of Managers on Password Strength and Reuse

Despite their well-known security problems, passwords are still the incumbent authentication method for virtually all online services. To remedy the situation, end-users are very often referred to password managers as a solution to the password reuse and password weakness problems. However, to date the actual impact of password managers on password security and reuse has not been studied systematically. In this paper, we provide the first large-scale study of the password managers' influence on users' real-life passwords. From 476 participants of an online survey on users' password creation and management strategies, we recruit 170 participants that allowed us to monitor their passwords in-situ through a browser plugin. In contrast to prior work, we collect the passwords' entry methods (e.g., human or password manager) in addition to the passwords and their metrics. Based on our collected data and our survey, we gain a more complete picture of the factors that influence our participants' passwords' strength and reuse. We quantify for the first time that password managers indeed benefit the password strength and uniqueness, however, also our results also suggest that those benefits depend on the users' strategies and that managers without password generators rather aggravate the existing problems.

cs.CR↗