Mogens Dalgaard

Mogens Dalgaard

Auto-generated profile, based on publicly available information

Mogens Dalgaard

Department of Physics and Astronomy, Aarhus University, Ny Munkegade 120, 8000 Aarhus C, Denmark

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 January 2020 in npj Quantum Information
Verified
Journal Article
Open access
PDF file
MDMogens Dalgaard
Mogens Dalgaard
FMFelix Motzoi
Felix Motzoi
JSJens Jakob Sørensen
Jens Jakob Sørensen
JSJacob Sherson
Jacob Sherson
While a large number of algorithms for optimizing quantum dynamics for different objectives have been developed, a common limitation is the reliance on good initial guesses, being either random or based on heuristics and intuitions. Here we implement a tabula rasa deep quantum exploration version of the Deepmind AlphaZero algorithm for systematically averting this limitation. AlphaZero employs a deep neural network in conjunction with deep lookahead in a guided tree search, which allows for predictive hidden-variable approximation of the quantum parameter landscape. To emphasize transferability, we apply and benchmark the algorithm on three classes of control problems using only a single common set of algorithmic hyperparameters. AlphaZero achieves substantial improvements in both the quality and quantity of good solution clusters compared to earlier methods. It is able to spontaneously learn unexpected hidden structure and global symmetry in the solutions, going beyond even human heuristics.

21 October 2020 in Physical Review A
Verified
Journal Article
Open access
PDF file
MDMogens Dalgaard
Mogens Dalgaard
FMFelix Motzoi
Felix Motzoi
JJJesper Hasseriis Mohr Jensen
Jesper Hasseriis Mohr Jensen
JSJacob Sherson
Jacob Sherson
Efficient optimization of quantum systems is a necessity for reaching fault-tolerant thresholds. A standard tool for optimizing simulated quantum dynamics is the gradient-based grape algorithm, which has been successfully applied in a wide range of different branches of quantum physics. In this work, we derive and implement exact second-order analytical derivatives of the coherent dynamics and find improvements compared to the standard of optimizing with the approximate second-order Broyden-Fletcher-Goldfarb-Shanno algorithm. We demonstrate performance improvements for both the best and the average errors of constrained unitary gate synthesis on a circuit-QED system over a broad range of different gate durations.

3 January 2022 in Physical Review A
Verified
Journal Article
Closed access
PDF file
MDMogens Dalgaard
Mogens Dalgaard
FMFelix Motzoi
Felix Motzoi
JSJacob Sherson
Jacob Sherson
State-of-the-art quantum algorithms routinely tune dynamically parametrized cost functionals for combinatorics, machine learning, equation solving, and energy minimization. However, large search complexity often demands many (noisy) quantum measurements, leading to the increasing use of classical probability models to estimate which areas in the cost functional landscape are of highest interest. Introducing deep learning based modeling of the landscape, we demonstrate order-of-magnitude increases in accuracy and speed over state-of-the-art Bayesian methods with respect to control of a many-body spin chain. Moreover, once trained, the deep neural network enables the extraction of information at a much faster rate than conventional numerical simulation. This allows for on-the-fly experimental optimizations of circuits with large depth and moderate width and detailed classification of problem complexity and navigability throughout the phase diagram of the landscape.

14 December 2018 in Physical Review A
Verified
Journal Article
Closed access
PDF file
JSJ. J. W. H. Sørensen
J. J. W. H. Sørensen
MDM. Dalgaard
M. Dalgaard
AKA. H. Kiilerich
A. H. Kiilerich
KMK. Mølmer
K. Mølmer
JSJ. F. Sherson
J. F. Sherson
We introduce an efficient iterative method to prepare a target state in Hilbert spaces with high dimensionality using a combination of unitary evolution, measurements, and quantum Zeno dynamics. The latter confines the evolution within Zeno subspaces of decreasing size. This gives an exponential speedup relative to the case of states evolving in the full Hilbert space between projective measurements. We demonstrate our approach on the control problem of rapidly transferring a superfluid into the Mott insulator in the Bose-Hubbard model. We discuss the general applicability of the method by preparing arbitrary superpositions with random Hamiltonians.

13 April 2022 in Physical Review Letters
Verified
Journal Article
Closed access
PDF file
MDMogens Dalgaard
Mogens Dalgaard
CWCarrie A. Weidner
Carrie A. Weidner
FMFelix Motzoi
Felix Motzoi
Many quantum technologies rely on high-precision dynamics, which raises the question of how these are influenced by the experimental uncertainties that are always present in real-life settings. A standard approach in the literature to assess this is Monte Carlo sampling, which suffers from two major drawbacks. First, it is computationally expensive. Second, it does not reveal the effect that each individual uncertainty parameter has on the state of the system. In this Letter, we evade both these drawbacks by incorporating propagation of uncertainty directly into simulations of quantum dynamics, thereby obtaining a method that is orders of magnitude faster than Monte Carlo simulations and directly provides information on how each uncertainty parameter influences the system dynamics. Additionally, we compare our method to experimental results obtained using the IBM quantum computers.