Abstract This work introduces a novel hybrid quantum-classical machine learning framework in which a Reinforcement Learning (RL) agent autonomously constructs quantum circuits to optimize kernels for a Quantum Support Vector Classifier (QSVC). Unlike traditional approaches that rely on fixed or manually designed quantum feature maps, the proposed method employs a Deep Q-Network (DQN) agent to dynamically select quantum gates and their parameters. These configurations define a parameterized quantum circuit whose fidelity-based kernel is used to perform classification tasks. The RL agent is trained to maximize classification accuracy by interacting with an environment that evaluates circuit performance. Experiments conducted on a realistic and potentially biased educational dataset demonstrate that the RL-optimized QSVC outperforms both classical SVC and standard QSVC baselines in terms of accuracy and fairness. Furthermore, SHAP-based explainability analysis reveals a reduction in model bias, suggesting improved interpretability and robustness. This study highlights the potential of reinforcement learning for automated quantum circuit design and kernel optimization in quantum machine learning.
Abstract This work introduces a novel hybrid quantum-classical machine learning framework in which a Reinforcement Learning (RL) agent autonomously constructs quantum circuits to optimize kernels for a Quantum Support Vector Classifier (QSVC). Unlike traditional approaches that rely on fixed or manually designed quantum feature maps, the proposed method employs a Deep Q-Network (DQN) agent to dynamically select quantum gates and their parameters. These configurations define a parameterized quantum circuit whose fidelity-based kernel is used to perform classification tasks. The RL agent is trained to maximize classification accuracy by interacting with an environment that evaluates circuit performance. Experiments conducted on a realistic and potentially biased educational dataset demonstrate that the RL-optimized QSVC outperforms both classical SVC and standard QSVC baselines in terms of accuracy and fairness. Furthermore, SHAP-based explainability analysis reveals a reduction in model bias, suggesting improved interpretability and robustness. This study highlights the potential of reinforcement learning for automated quantum circuit design and kernel optimization in quantum machine learning.
Quantum machine learning has emerged as a promising frontier, leveraging the principles of quantum computing to enhance classical algorithms. This paper proposes an optimized Quantum Support Vector Classifier (QSVC) for the classification of the MNIST dataset (a benchmark in handwritten digit recognition). We employ a Chebyshev Parameterized Quantum Circuit (PQC) with an increased number of qubits and layers, and utilize an existing projected quantum kernel method to effectively capture the complex patterns in the data. To enhance model robustness, data augmentation techniques are applied, and dimensionality reduction is performed using Principal Component Analysis (PCA). The proposed QSVC demonstrates competitive accuracy on both the training and testing datasets, highlighting the potential of quantum kernel methods in practical machine learning applications. A comparative analysis with classical Support Vector Machines (SVM) demonstrates the feasibility of quantum methods for high-dimensional classification tasks. Additionally, we investigate the impact of varying the number of qubits in the quantum circuit, providing in-depth insights into the scalability and performance trade-offs of quantum kernel methods.
This paper presents a novel approach to quantum circuit compilation leveraging reinforcement learning (RL). The core idea is to train an RL agent to dynamically optimize quantum circuit structures, aiming to minimize gate count and enhance overall circuit performance. Traditional quantum circuit compilation methods often rely on static heuristics and pre-defined optimization strategies, which may not be optimal for all quantum algorithms or hardware platforms. Our approach addresses this limitation by allowing the RL agent to learn a policy that adapts to the specific characteristics of the algorithm and the target quantum hardware. The agent learns by interacting with a simulated quantum circuit environment, receiving rewards based on circuit fidelity, depth, and gate count. The results demonstrate the potential of RL to significantly improve the efficiency and robustness of quantum computations. The key components of the system are: a high-level quantum algorithm representation, an RL agent, a quantum circuit simulator, and a reward function. The agent's policy is represented as a neural network, enabling it to generalize across different algorithms and hardware. We demonstrate the effectiveness of this approach through simulations, showcasing its ability to outperform conventional compilation methods in terms of reducing circuit depth and gate count while maintaining high fidelity.
Quantum circuit compilation is a crucial step in realizing quantum algorithms on real quantum hardware. Traditionally, compilation methods are largely static, relying on pre-defined mappings and optimizations that fail to account for the inherent noise and variability present in quantum devices. This paper proposes a novel approach leveraging Reinforcement Learning (RL) to dynamically generate and optimize quantum circuits. An RL agent is trained to learn a policy for circuit design, adapting to specific hardware characteristics and real-time measurement data. The agent's goal is to minimize the overall error rate of the compiled circuit. This dynamic compilation framework offers a feedback loop between hardware performance and circuit design, resulting in adaptive and robust quantum computation. The core claim of this work is that traditional static compilation methods are insufficient, and dynamic compilation driven by RL offers a superior solution. The approach presented here represents a significant advancement towards realizing the full potential of quantum computing by mitigating the effects of hardware imperfections. The framework includes a detailed description of the RL agent architecture, the reward function design, and the exploration strategy employed. The effectiveness of the proposed method is demonstrated through simulations, highlighting its ability to outperform static compilation strategies in noisy environments.
Quantum computing offers a powerful new paradigm for enhancing classical machine learning, particularly in classification tasks. In this paper, we examine the use of quantum kernels within Support Vector Classification (SVC) and compare their performance to classical SVC. Our experiments focus on the Iris and Wine datasets. For the Wine dataset, we applied Principal Component Analysis (PCA), reducing the feature space from 13 to 4 dimensions to improve computational efficiency. The results demonstrate that Quantum-enhanced SVC (QSVC) outperforms classical SVC, achieving an accuracy improvement from 96.15% with classical SVM to 98.05% with QSVC for the Wine dataset. Similarly, for the Iris dataset, classical SVC achieves near-perfect accuracy at 100%, while QSVC closely follows with 97.25% accuracy. These results highlight the capability of quantum kernels to improve both accuracy and efficiency in classification tasks, providing a promising direction for scalable quantum-classical hybrid systems in machine learning.
One of the most important initial steps in quantum computing is high-fidelity quantum state preparation, because errors in it can affect the final computational result. Therefore, one of the major challenges is to design a quantum circuit that can achieve the desired target state accurately with a smaller number of gates. Traditionally, circuits are designed using predefined rules and mathematical decompositions, but this makes it difficult to design circuits for different and complex target states. This paper presents TATVA, a reinforcement learning (RL) system for synthesizing quantum circuits. TATVA views circuit synthesis as a sequential decision problem in which the agents select one gate at a time and calculate the fidelity after each applied gate, with the gate producing the highest fidelity selected for the circuit. The system introduces a parallel architecture that uses two RL agents, Deep Q-Network (DQN) and Proximal Policy Optimization (PPO), along with Qiskit's statevector simulator. The system can synthesize up to 5 qubits while achieving a fidelity of 0.999999 and producing compact circuits. After achieving the desired fidelity, it performs a post-hoc optimization step that reduces the circuit depth while preserving the achieved fidelity. The circuit is finalized using fidelity, success rate, gate count, and circuit depth. Target states that are not used during training are used to test whether the system has learned effectively and can generalize beyond the training states. This approach aims to enable automatic quantum circuit synthesis with high fidelity using reinforcement learning.
Quantum computing promises significant computational advantages over classical methods in certain applications but is currently too unstable for executing complex tasks. High-fidelity quantum circuits are critical to overcoming this challenge, yet they are computationally challenging to design. On current noisy intermediate-scale quantum (NISQ) hardware, shallower circuits with fewer two-qubit gates generally run with higher fidelity. Increased fidelity can be sought by sequentially rewriting the circuit, which involves a vast search space. This costly search space exploration could be circumvented by using reinforcement learning (RL) to teach an agent to rewrite the circuit in a sequential decision-making process. This thesis aims to implement a Deep Q-Network (DQN) algorithm to optimize BernsteinVazirani (BVZ) circuits. The DQN algorithm interacts with a Cirq-based environment. The environment provides valid rewrite actions to the DQN algorithm, which gains a scalar reward emphasizing depth and gate reduction. In the thesis we compare two different state encodings as input for the neural network: a compact state summarizing vector representation with three features, and an image-like encoding that maps circuit state topology with tokens through a fixed embedding. Both input encodings use the same action rewrite set and reward design. The network is implemented in PyTorch and trained with Adam optimizer, and runs on GPU via CUDA. Empirically, the agent shows promising initial optimizing performance. On several BVZ circuits with varying qubit count and topology, the agent drives the depth toward the theoretical optimum of three. The simplistic vector encoding already reaches reasonable circuit depth plateaus, with the image-like encoding together with hyperparameter tuning improving reliability, showing avenues for continued improvement. We conclude that DQN is a viable foundation for automated circuit optimization on BVZ circuits. To achieve reliable circuit optimization at scale, future research should investigate even richer encoding methods and reward function shaping. Finally, the results support applying the DQN to be applied to circuit families beyond BVZ.