Mode connectivity in the loss landscape of parameterized quantum circuits

Hamilton, Kathleen E.; Lynn, Emily; Pooser, Raphael C.

doi:10.1007/s42484-021-00059-5

Mode connectivity in the loss landscape of parameterized quantum circuits

Journal Article · 11 May 2022 · Quantum Machine Intelligence

DOI:https://doi.org/10.1007/s42484-021-00059-5· OSTI ID:1885380

^[1]; ^[2]; ^[1]

Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Taylor University, Upland, IN (United States)

Variational training of parameterized quantum circuits (PQCs) underpins many workflows employed on near-term noisy intermediate scale quantum (NISQ) devices. It is a hybrid quantum-classical approach that minimizes an associated cost function in order to train a parameterized ansatz. In this work we adapt the qualitative loss landscape characterization for neural networks introduced in Goodfellow et al. (2014); Li et al. (2017) and tests for connectivity used in Draxler et al. (2018) to study the loss landscape features in PQC training. We present results for PQCs trained on a simple regression task, using the bilayer circuit ansatz, which consists of alternating layers of parameterized rotation gates and entangling gates. Multiple circuits are trained with 3 different batch gradient optimizers: stochastic gradient descent, the quantum natural gradient, and Adam. We identify large features in the landscape that can lead to faster convergence in training workflows.

View Accepted Manuscript (DOE)

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); USDOE Office of Science (SC), Workforce Development for Teachers and Scientists (WDTS)

Grant/Contract Number:: AC05-00OR22725

OSTI ID:: 1885380

Journal Information:: Quantum Machine Intelligence, Journal Name: Quantum Machine Intelligence Journal Issue: 1 Vol. 4; ISSN 2524-4906

Publisher:: Springer NatureCopyright Statement

Country of Publication:: United States

Language:: English

References (26)

Flat Minima Hochreiter, Sepp; Schmidhuber, Jürgen Neural Computation, Vol. 9, Issue 1 https://doi.org/10.1162/neco.1997.9.1.1	journal	January 1997
Barren plateaus in quantum neural network training landscapes McClean, Jarrod R.; Boixo, Sergio; Smelyanskiy, Vadim N. Nature Communications, Vol. 9, Issue 1 https://doi.org/10.1038/s41467-018-07090-4	journal	November 2018
Data re-uploading for a universal quantum classifier Pérez-Salinas, Adrián; Cervera-Lierta, Alba; Gil-Fuster, Elies Quantum, Vol. 4 https://doi.org/10.22331/q-2020-02-06-226	journal	February 2020
Mean shift: a robust approach toward feature space analysis Comaniciu, D.; Meer, P. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, Issue 5 https://doi.org/10.1109/34.1000236	journal	May 2002
Cost function dependent barren plateaus in shallow parametrized quantum circuits Cerezo, M.; Sone, Akira; Volkoff, Tyler Nature Communications, Vol. 12, Issue 1 https://doi.org/10.1038/s41467-021-21728-w	journal	March 2021
Archetypal energy landscapes Wales, David J.; Miller, Mark A.; Walsh, Tiffany R. Nature, Vol. 394, Issue 6695 https://doi.org/10.1038/29487	journal	August 1998
Connecting Ansatz Expressibility to Gradient Magnitudes and Barren Plateaus Holmes, Zoë; Sharma, Kunal; Cerezo, M. PRX Quantum, Vol. 3, Issue 1 https://doi.org/10.1103/PRXQuantum.3.010313	journal	January 2022
Evaluating analytic gradients on quantum hardware Schuld, Maria; Bergholm, Ville; Gogolin, Christian Physical Review A, Vol. 99, Issue 3 https://doi.org/10.1103/PhysRevA.99.032331	journal	March 2019
Parameterized quantum circuits as machine learning models Benedetti, Marcello; Lloyd, Erika; Sack, Stefan Quantum Science and Technology, Vol. 4, Issue 4 https://doi.org/10.1088/2058-9565/ab4eb5	journal	October 2019
Low-Depth Gradient Measurements Can Improve Convergence in Variational Hybrid Quantum-Classical Algorithms Harrow, Aram W.; Napp, John C. Physical Review Letters, Vol. 126, Issue 14 https://doi.org/10.1103/PhysRevLett.126.140502	journal	April 2021
Glassy Nature of the Hard Phase in Inference Problems Antenucci, Fabrizio; Franz, Silvio; Urbani, Pierfrancesco Physical Review X, Vol. 9, Issue 1 https://doi.org/10.1103/PhysRevX.9.011020	journal	January 2019
Shaping the learning landscape in neural networks around wide flat minima Baldassi, Carlo; Pittorino, Fabrizio; Zecchina, Riccardo Proceedings of the National Academy of Sciences, Vol. 117, Issue 1 https://doi.org/10.1073/pnas.1908636117	journal	December 2019
Effect of data encoding on the expressive power of variational quantum-machine-learning models Schuld, Maria; Sweke, Ryan; Meyer, Johannes Jakob Physical Review A, Vol. 103, Issue 3 https://doi.org/10.1103/PhysRevA.103.032430	journal	March 2021
Noise-induced barren plateaus in variational quantum algorithms Wang, Samson; Fontana, Enrico; Cerezo, M. Nature Communications, Vol. 12, Issue 1 https://doi.org/10.1038/s41467-021-27045-6	journal	November 2021
Estimating the gradient and higher-order derivatives on quantum hardware Mari, Andrea; Bromley, Thomas R.; Killoran, Nathan Physical Review A, Vol. 103, Issue 1 https://doi.org/10.1103/PhysRevA.103.012405	journal	January 2021
Characterizing the loss landscape of variational quantum circuits Huembeli, Patrick; Dauphin, Alexandre Quantum Science and Technology, Vol. 6, Issue 2 https://doi.org/10.1088/2058-9565/abdbc9	journal	February 2021
Quantum control landscape for a two-level system near the quantum speed limit Larocca, Martín; Poggi, Pablo M.; Wisniacki, Diego A. Journal of Physics A: Mathematical and Theoretical, Vol. 51, Issue 38 https://doi.org/10.1088/1751-8121/aad657	journal	August 2018
Quantum Natural Gradient Stokes, James; Izaac, Josh; Killoran, Nathan Quantum, Vol. 4 https://doi.org/10.22331/q-2020-05-25-269	journal	May 2020
Expressibility and Entangling Capability of Parameterized Quantum Circuits for Hybrid Quantum‐Classical Algorithms Sim, Sukin; Johnson, Peter D.; Aspuru‐Guzik, Alán Advanced Quantum Technologies, Vol. 2, Issue 12 https://doi.org/10.1002/qute.201900070	journal	October 2019
The power of quantum neural networks Abbas, Amira; Sutter, David; Zoufal, Christa Nature Computational Science, Vol. 1, Issue 6 https://doi.org/10.1038/s43588-021-00084-1	journal	June 2021
Comparing dynamics: deep neural networks versus glassy systems Baity-Jesi, Marco; Sagun, Levent; Geiger, Mario Journal of Statistical Mechanics: Theory and Experiment, Vol. 2019, Issue 12 https://doi.org/10.1088/1742-5468/ab3281	journal	December 2019
Equivalence of quantum barren plateaus to cost concentration and narrow gorges Arrasmith, Andrew; Holmes, Zoë; Cerezo, M. Quantum Science and Technology, Vol. 7, Issue 4 https://doi.org/10.1088/2058-9565/ac7d06	journal	August 2022
Quantum circuit learning Mitarai, K.; Negoro, M.; Kitagawa, M. Physical Review A, Vol. 98, Issue 3 https://doi.org/10.1103/PhysRevA.98.032309	journal	September 2018
Control landscapes are almost always trap free: a geometric assessment Russell, Benjamin; Rabitz, Herschel; Wu, Re-Bing Journal of Physics A: Mathematical and Theoretical, Vol. 50, Issue 20 https://doi.org/10.1088/1751-8121/aa6b77	journal	April 2017
Improved tangent estimate in the nudged elastic band method for finding minimum energy paths and saddle points Henkelman, Graeme; Jónsson, Hannes The Journal of Chemical Physics, Vol. 113, Issue 22 https://doi.org/10.1063/1.1323224	journal	December 2000
Supervised learning with quantum-enhanced feature spaces Havlíček, Vojtěch; Córcoles, Antonio D.; Temme, Kristan Nature, Vol. 567, Issue 7747 https://doi.org/10.1038/s41586-019-0980-2	journal	March 2019

Similar Records

Graph Neural Networks for Parameterized Quantum Circuits Expressibility Estimation (Rev.1)

Technical Report · 2024 · OSTI ID:2350603

Aktar, Shamminuj; Baertschi, Andreas; Badawy, Abdel-Hameed A.; +2 more

Large gradients via correlation in random parameterized quantum circuits

Journal Article · 2021 · Quantum Science and Technology · OSTI ID:1764966

Volkoff, Tyler James; Coles, Patrick Joseph

Mode connectivity in the QCBM loss landscape: ICCAD Special Session Paper

Conference · 2021 · OSTI ID:2538511

Hamilton, Kathleen; Lynn, Emily; Leyton Ortega, Vicente; +2 more

Related Subjects

97 MATHEMATICS AND COMPUTING
quantum machine learning
gradient based optimization
visualization
circuit design
mode connectivity

Mode connectivity in the loss landscape of parameterized quantum circuits

Citation Formats

References (26)

Similar Records

Related Subjects