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

arXiv:2506.11565 · Tenth International Conference on Information Optics and Photonics

Gradients of unitary optical neural networks using parameter-shift rule

Jinzhe Jiang, Xin Zhang, Li Chen, Yunlong Yu, Hailing Liu

Published Dec 8, 2025• 1 citation
arXivHTMLPDF

Abstract

This paper explores the application of the parameter-shift rule (PSR) for computing gradients in unitary optical neural networks (UONNs). While backpropagation has been fundamental to training conventional neural networks, its implementation in optical neural networks faces significant challenges due to the physical constraints of optical systems. We demonstrate how PSR, which calculates gradients by evaluating functions at shifted parameter values, can be effectively adapted for training UONNs constructed from Mach-Zehnder interferometer meshes. We develop a theoretical framework for two distinct scenarios, real-valued outputs and complex-valued outputs of optical networks, which derives exact analytical gradients directly from hardware measurements for phase parameters in optical neural networks. For the first real-valued scenario, the gradient calculation directly conforms to PSR. For the second complex-valued scenario, we improved PSR to obtain the gradient. Then we simulate the training of optical neural networks based on optical PSR and both scenarios complete the classification tasks well. The optical PSR offers a powerful and analytically precise approach for calculating gradients in unitary optical neural networks. This approach offers a promising alternative to traditional in silico training methods and circumvents the limitations of both finite difference approximations and all-optical backpropagation implementations. We present the theoretical framework and practical methodology for applying PSR to optimize phase parameters in optical neural networks, potentially advancing the development of efficient hardware-based training strategies for optical computing systems.

DOI
10.1117/12.3085980

Related papers

Efficient training of unitary optical neural networks

Kebang Lu, Xianxin Guo

Deep learning has profoundly reshaped the technology landscape in numerous scientific areas and industrial sectors. This technology advancement is, nevertheless, confronted with severe bottlenecks in digital computing. Optical neural network presents a promising solution due to the ultra-high computing speed and energy efficiency. In this work, we present systematic study of unitary optical neural network (UONN) as an approach towards optical deep learning. Our results show that the UONN can be trained to high accuracy through special unitary gradient descent optimization, and the UONN is robust against physical imperfections and noises, hence it is more suitable for physical implementation than existing ONNs.

Optics Express•Oct 27, 2023•7 citations

Efficient Parameter-Shift Rule Implementation for Computing Gradient on Quantum Simulators

Vu Tuan Hai, Vu Trung Duong Le, Pham Hoai Luan +1 more

Quantum computing is an active research interest in the new computational area with various applications. In the quantum machine learning field, the parameterized quantum circuit is a core learnable model; this model is updated iteratively by the general Parameter-Shift Rule (PSR) technique. In many qubits and parameter scenarios, using PSR consumes significant computational resources, particularly in the quantum simulator. Therefore, this research proposes a method to perform the PSR more efficiently for matrix multiplication and state-based real-world quantum simulators. The results show that our method is faster than baseline matrix multiplication-based 12 times and baseline state-based 39 times on average experiment.

International Conference on Advanced Technologies for Communications•Oct 17, 2024•1 citation

Using a single circuit to compute the gradients with respect to all parameters of a quantum neural network

Guang Ping He

Finding gradients is a crucial step in training machine learning models. For quantum neural networks, computing gradients using the parameter-shift rule requires calculating the cost function twice for each adjustable parameter in the network. When the total number of parameters is large, the quantum circuit must be repeatedly adjusted and executed, leading to significant computational overhead. Here, we propose an approach to compute all gradients using only a single circuit, significantly reducing both the circuit depth and the number of classical registers required. Although the theoretical complexity of gradient calculation and the total number of measurements are not reduced by this approach, a considerable reduction in the number of unique circuit compilations and job submissions is achieved. We experimentally validate our approach on both quantum simulators and IBM’s real quantum hardware, demonstrating that our method significantly reduces circuit compilation time compared to the conventional approach, resulting in a substantial speedup in total runtime.

arXiv:2307.08167•Frontiers of Physics•Jan 1, 2026•6 citations

Single-component gradient rules for variational quantum algorithms

Thomas Hubregtsen, Frederik Wilde, Shozab Qasim +1 more

Abstract Many near-term quantum computing algorithms are conceived as variational quantum algorithms, in which parameterized quantum circuits are optimized in a hybrid quantum–classical setup. Examples are variational quantum eigensolvers, quantum approximate optimization algorithms as well as various algorithms in the context of quantum-assisted machine learning. A common bottleneck of any such algorithm is constituted by the optimization of the variational parameters. A popular set of optimization methods work on the estimate of the gradient, obtained by means of circuit evaluations. We will refer to the way in which one can combine these circuit evaluations as gradient rules. This work provides a comprehensive picture of the family of gradient rules that vary parameters of quantum gates individually. The most prominent known members of this family are the parameter shift rule (PSR) and the finite differences method. To unite this family, we propose a generalized PSR that expresses all members of the aforementioned family as special cases, and discuss how all of these can be seen as providing access to a linear combination of exact first- and second-order derivatives. We further prove that a PSR with one non-shifted evaluation and only one shifted circuit evaluation does not exist, and introduce a novel perspective for approaching new gradient rules.

arXiv:2106.01388•Quantum Science and Technology•Apr 19, 2022•23 citations

Photonic parameter-shift rule: Enabling gradient computation for photonic quantum computers

Axel Pappalardo, Pierre-Emmanuel Emeriau, Giovanni de Felice +5 more

We present a method for gradient computation in quantum algorithms implemented on linear optical quantum computing platforms. While parameter-shift rules have become a staple in qubit gate-based quantum computing for calculating gradients, their direct application to photonic platforms has been hindered by the infinite dimensionality of phase-shift operators in Fock space. We introduce a photonic parameter-shift rule that overcomes this limitation, providing an exact formula for gradient computation in linear optical quantum processors. Our method scales linearly with the number of input photons and utilizes the same parametrized photonic circuit with shifted parameters for each evaluation. This advancement bridges a crucial gap in photonic quantum computing, enabling efficient gradient-based optimization for variational quantum algorithms on near-term photonic quantum processors. We demonstrate the efficacy of our approach through numerical simulations in quantum chemistry and generative modeling tasks, showing superior optimization performance as well as robustness to noise from finite sampling and photon distinguishability compared to other gradient-based and gradient-free methods.

arXiv:2410.02726•Physical Review A•Mar 25, 2025•16 citations

Exact gradients for linear optics with single photons

Giorgio Facelli, Roberts, David D., Hugo Wallner +3 more

Though parameter shift rules have drastically improved gradient estimation methods for several types of quantum circuits, leading to improved performance in downstream tasks, so far they have not been transferable to linear optics with single photons. In this work, we derive an analytical formula for the gradients in these circuits with respect to phaseshifters via a generalized parameter shift rule, where the number of parameter shifts depends linearly on the total number of photons. Experimentally, this enables access to derivatives in photonic systems without the need for finite difference approximations. Building on this, we propose two strategies through which one can reduce the number of shifts in the expression, and hence reduce the overall sample complexity. Numerically, we show that this generalized parameter-shift rule can converge to the minimum of a cost function with fewer parameter update steps than alternative techniques. We anticipate that this method will open up new avenues to solving optimization problems with photonic systems, as well as provide new techniques for the experimental characterization and control of linear optical systems.

arXiv:2409.16369•Sep 24, 2024•25 citations
This paper explores the application of the parameter-shift rule (PSR) for computing gradients in unitary optical neural networks (UONNs). While backpropagation has been fundamental to training conventional neural networks, its implementation in optical neural networks faces significant challenges due to the physical constraints of optical systems. We demonstrate how PSR, which calculates gradients by evaluating functions at shifted parameter values, can be effectively adapted for training UONNs constructed from Mach-Zehnder interferometer meshes. We develop a theoretical framework for two distinct scenarios, real-valued outputs and complex-valued outputs of optical networks, which derives exact analytical gradients directly from hardware measurements for phase parameters in optical neural networks. For the first real-valued scenario, the gradient calculation directly conforms to PSR. For the second complex-valued scenario, we improved PSR to obtain the gradient. Then we simulate the training of optical neural networks based on optical PSR and both scenarios complete the classification tasks well. The optical PSR offers a powerful and analytically precise approach for calculating gradients in unitary optical neural networks. This approach offers a promising alternative to traditional in silico training methods and circumvents the limitations of both finite difference approximations and all-optical backpropagation implementations. We present the theoretical framework and practical methodology for applying PSR to optimize phase parameters in optical neural networks, potentially advancing the development of efficient hardware-based training strategies for optical computing systems.
Deep learning has profoundly reshaped the technology landscape in numerous scientific areas and industrial sectors. This technology advancement is, nevertheless, confronted with severe bottlenecks in digital computing. Optical neural network presents a promising solution due to the ultra-high computing speed and energy efficiency. In this work, we present systematic study of unitary optical neural network (UONN) as an approach towards optical deep learning. Our results show that the UONN can be trained to high accuracy through special unitary gradient descent optimization, and the UONN is robust against physical imperfections and noises, hence it is more suitable for physical implementation than existing ONNs.
Quantum computing is an active research interest in the new computational area with various applications. In the quantum machine learning field, the parameterized quantum circuit is a core learnable model; this model is updated iteratively by the general Parameter-Shift Rule (PSR) technique. In many qubits and parameter scenarios, using PSR consumes significant computational resources, particularly in the quantum simulator. Therefore, this research proposes a method to perform the PSR more efficiently for matrix multiplication and state-based real-world quantum simulators. The results show that our method is faster than baseline matrix multiplication-based 12 times and baseline state-based 39 times on average experiment.
Finding gradients is a crucial step in training machine learning models. For quantum neural networks, computing gradients using the parameter-shift rule requires calculating the cost function twice for each adjustable parameter in the network. When the total number of parameters is large, the quantum circuit must be repeatedly adjusted and executed, leading to significant computational overhead. Here, we propose an approach to compute all gradients using only a single circuit, significantly reducing both the circuit depth and the number of classical registers required. Although the theoretical complexity of gradient calculation and the total number of measurements are not reduced by this approach, a considerable reduction in the number of unique circuit compilations and job submissions is achieved. We experimentally validate our approach on both quantum simulators and IBM’s real quantum hardware, demonstrating that our method significantly reduces circuit compilation time compared to the conventional approach, resulting in a substantial speedup in total runtime.
Abstract Many near-term quantum computing algorithms are conceived as variational quantum algorithms, in which parameterized quantum circuits are optimized in a hybrid quantum–classical setup. Examples are variational quantum eigensolvers, quantum approximate optimization algorithms as well as various algorithms in the context of quantum-assisted machine learning. A common bottleneck of any such algorithm is constituted by the optimization of the variational parameters. A popular set of optimization methods work on the estimate of the gradient, obtained by means of circuit evaluations. We will refer to the way in which one can combine these circuit evaluations as gradient rules. This work provides a comprehensive picture of the family of gradient rules that vary parameters of quantum gates individually. The most prominent known members of this family are the parameter shift rule (PSR) and the finite differences method. To unite this family, we propose a generalized PSR that expresses all members of the aforementioned family as special cases, and discuss how all of these can be seen as providing access to a linear combination of exact first- and second-order derivatives. We further prove that a PSR with one non-shifted evaluation and only one shifted circuit evaluation does not exist, and introduce a novel perspective for approaching new gradient rules.
We present a method for gradient computation in quantum algorithms implemented on linear optical quantum computing platforms. While parameter-shift rules have become a staple in qubit gate-based quantum computing for calculating gradients, their direct application to photonic platforms has been hindered by the infinite dimensionality of phase-shift operators in Fock space. We introduce a photonic parameter-shift rule that overcomes this limitation, providing an exact formula for gradient computation in linear optical quantum processors. Our method scales linearly with the number of input photons and utilizes the same parametrized photonic circuit with shifted parameters for each evaluation. This advancement bridges a crucial gap in photonic quantum computing, enabling efficient gradient-based optimization for variational quantum algorithms on near-term photonic quantum processors. We demonstrate the efficacy of our approach through numerical simulations in quantum chemistry and generative modeling tasks, showing superior optimization performance as well as robustness to noise from finite sampling and photon distinguishability compared to other gradient-based and gradient-free methods.
Though parameter shift rules have drastically improved gradient estimation methods for several types of quantum circuits, leading to improved performance in downstream tasks, so far they have not been transferable to linear optics with single photons. In this work, we derive an analytical formula for the gradients in these circuits with respect to phaseshifters via a generalized parameter shift rule, where the number of parameter shifts depends linearly on the total number of photons. Experimentally, this enables access to derivatives in photonic systems without the need for finite difference approximations. Building on this, we propose two strategies through which one can reduce the number of shifts in the expression, and hence reduce the overall sample complexity. Numerically, we show that this generalized parameter-shift rule can converge to the minimum of a cost function with fewer parameter update steps than alternative techniques. We anticipate that this method will open up new avenues to solving optimization problems with photonic systems, as well as provide new techniques for the experimental characterization and control of linear optical systems.