Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices

Xian Wang; Paul Kairys; Sri Hari Krishna Narayanan; Jan Huckelheim; Paul Hovland

doi:10.1109/QCS56647.2022.00016

Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices

Xian Wang, Paul Kairys, Sri Hari Krishna Narayanan, Jan Huckelheim, Paul Hovland

Quantum Computational Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

Quantum optimal control problems are typically solved by gradient-based algorithms such as GRAPE, which suffer from exponential growth in storage with increasing number of qubits and linear growth in memory requirements with increasing number of time steps. Employing QOC for discrete lattices reveals that these memory requirements are a barrier for simulating large models or long time spans. We employ a non-standard differentiable programming approach that significantly reduces the memory requirements at the cost of a reasonable amount of recomputation. The approach exploits invertibility properties of the unitary matrices to reverse the computation during back-propagation. We utilize QOC software written in the differentiable programming framework JAX that implements this approach, and demonstrate its effectiveness for lattice gauge theory.

Original language	English
Title of host publication	Proceedings of QCS 2022
Subtitle of host publication	3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	94-99
Number of pages	6
ISBN (Electronic)	9781665475365
DOIs	https://doi.org/10.1109/QCS56647.2022.00016
State	Published - 2022
Event	3rd IEEE/ACM International Workshop on Quantum Computing Software, QCS 2022 - Dallas, United States Duration: Nov 13 2022 → Nov 13 2022

Publication series

Name	Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis

Conference

Conference	3rd IEEE/ACM International Workshop on Quantum Computing Software, QCS 2022
Country/Territory	United States
City	Dallas
Period	11/13/22 → 11/13/22

Funding

This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, under the Accelerated Research in Quantum Computing and Applied Mathematics programs, under contract DE-AC02-06CH11357, and by the National Science Foundation Mathematical Sciences Graduate Internship. We gratefully acknowledge the computing resources provided on Bebop and Swing, a high-performance computing cluster operated by the Laboratory Computing Resource Center at Argonne National Laboratory.

Keywords

Automatic Differentiation
Lattice Gauge Theory
Quantum Optimal Control

Access to Document

10.1109/QCS56647.2022.00016

Cite this

Wang, X., Kairys, P., Narayanan, S. H. K., Huckelheim, J., & Hovland, P. (2022). Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices. In Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 94-99). (Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/QCS56647.2022.00016

Wang, Xian ; Kairys, Paul ; Narayanan, Sri Hari Krishna et al. / Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices. Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 94-99 (Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis).

@inproceedings{19aac401ee2d4d84bc7bfc6d5462707c,

title = "Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices",

abstract = "Quantum optimal control problems are typically solved by gradient-based algorithms such as GRAPE, which suffer from exponential growth in storage with increasing number of qubits and linear growth in memory requirements with increasing number of time steps. Employing QOC for discrete lattices reveals that these memory requirements are a barrier for simulating large models or long time spans. We employ a non-standard differentiable programming approach that significantly reduces the memory requirements at the cost of a reasonable amount of recomputation. The approach exploits invertibility properties of the unitary matrices to reverse the computation during back-propagation. We utilize QOC software written in the differentiable programming framework JAX that implements this approach, and demonstrate its effectiveness for lattice gauge theory.",

keywords = "Automatic Differentiation, Lattice Gauge Theory, Quantum Optimal Control",

author = "Xian Wang and Paul Kairys and Narayanan, {Sri Hari Krishna} and Jan Huckelheim and Paul Hovland",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 3rd IEEE/ACM International Workshop on Quantum Computing Software, QCS 2022 ; Conference date: 13-11-2022 Through 13-11-2022",

year = "2022",

doi = "10.1109/QCS56647.2022.00016",

language = "English",

series = "Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "94--99",

booktitle = "Proceedings of QCS 2022",

}

Wang, X, Kairys, P, Narayanan, SHK, Huckelheim, J & Hovland, P 2022, Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices. in Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis. Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis, Institute of Electrical and Electronics Engineers Inc., pp. 94-99, 3rd IEEE/ACM International Workshop on Quantum Computing Software, QCS 2022, Dallas, United States, 11/13/22. https://doi.org/10.1109/QCS56647.2022.00016

Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices. / Wang, Xian; Kairys, Paul; Narayanan, Sri Hari Krishna et al.
Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2022. p. 94-99 (Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices

AU - Wang, Xian

AU - Kairys, Paul

AU - Narayanan, Sri Hari Krishna

AU - Huckelheim, Jan

AU - Hovland, Paul

PY - 2022

Y1 - 2022

N2 - Quantum optimal control problems are typically solved by gradient-based algorithms such as GRAPE, which suffer from exponential growth in storage with increasing number of qubits and linear growth in memory requirements with increasing number of time steps. Employing QOC for discrete lattices reveals that these memory requirements are a barrier for simulating large models or long time spans. We employ a non-standard differentiable programming approach that significantly reduces the memory requirements at the cost of a reasonable amount of recomputation. The approach exploits invertibility properties of the unitary matrices to reverse the computation during back-propagation. We utilize QOC software written in the differentiable programming framework JAX that implements this approach, and demonstrate its effectiveness for lattice gauge theory.

AB - Quantum optimal control problems are typically solved by gradient-based algorithms such as GRAPE, which suffer from exponential growth in storage with increasing number of qubits and linear growth in memory requirements with increasing number of time steps. Employing QOC for discrete lattices reveals that these memory requirements are a barrier for simulating large models or long time spans. We employ a non-standard differentiable programming approach that significantly reduces the memory requirements at the cost of a reasonable amount of recomputation. The approach exploits invertibility properties of the unitary matrices to reverse the computation during back-propagation. We utilize QOC software written in the differentiable programming framework JAX that implements this approach, and demonstrate its effectiveness for lattice gauge theory.

KW - Automatic Differentiation

KW - Lattice Gauge Theory

KW - Quantum Optimal Control

UR - http://www.scopus.com/inward/record.url?scp=85148584945&partnerID=8YFLogxK

U2 - 10.1109/QCS56647.2022.00016

DO - 10.1109/QCS56647.2022.00016

M3 - Conference contribution

AN - SCOPUS:85148584945

T3 - Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis

SP - 94

EP - 99

BT - Proceedings of QCS 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd IEEE/ACM International Workshop on Quantum Computing Software, QCS 2022

Y2 - 13 November 2022 through 13 November 2022

ER -

Wang X, Kairys P, Narayanan SHK, Huckelheim J, Hovland P. Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices. In Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc. 2022. p. 94-99. (Proceedings of QCS 2022: 3rd International Workshop on Quantum Computing Software, Held in conjunction with SC 2022: The International Conference for High Performance Computing, Networking, Storage and Analysis). doi: 10.1109/QCS56647.2022.00016

Memory-Efficient Differentiable Programming for Quantum Optimal Control of Discrete Lattices

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this