In the era of noisy intermediate-scale quantum (NISQ) computing, Quantum Neural Networks (QNNs) have emerged as a promising approach for various applications, yet their training is often hindered by barren plateaus (BPs), where gradient variance vanishes exponentially as the qubit size increases.Most initializationbased mitigation strategies rely heavily on predesigned static parameter distributions, thereby lacking adaptability to diverse model sizes or data conditions.To address these limitations, we propose AdaInit, a foundational framework that leverages large language models with the submartingale property to iteratively synthesize initial parameters for QNNs that yield nonnegligible gradient variance, thereby mitigating BPs.Unlike conventional one-shot initialization methods, AdaInit adaptively explores the parameter space by incorporating dataset characteristics and gradient feedback, with theoretical guarantees of convergence to finding a set of effective initial parameters for QNNs.We provide rigorous theoretical analyses of the submartingale-based process and empirically validate that AdaInit consistently outperforms existing initialization methods in maintaining higher gradient variance across various QNN scales.We believe this work may initiate a new avenue to mitigate BPs.
In the era of noisy intermediate-scale quantum (NISQ) computing, Quantum Neural Networks (QNNs) have emerged as a promising approach for various applications, yet their training is often hindered by barren plateaus (BPs), where gradient variance vanishes exponentially as the qubit size increases.Most initializationbased mitigation strategies rely heavily on predesigned static parameter distributions, thereby lacking adaptability to diverse model sizes or data conditions.To address these limitations, we propose AdaInit, a foundational framework that leverages large language models with the submartingale property to iteratively synthesize initial parameters for QNNs that yield nonnegligible gradient variance, thereby mitigating BPs.Unlike conventional one-shot initialization methods, AdaInit adaptively explores the parameter space by incorporating dataset characteristics and gradient feedback, with theoretical guarantees of convergence to finding a set of effective initial parameters for QNNs.We provide rigorous theoretical analyses of the submartingale-based process and empirically validate that AdaInit consistently outperforms existing initialization methods in maintaining higher gradient variance across various QNN scales.We believe this work may initiate a new avenue to mitigate BPs.
Parameterized quantum circuits (PQCs) have emerged as a foundational element in the development and applications of quantum algorithms. However, when initialized with random parameter values, PQCs often exhibit barren plateaus (BP). These plateaus, characterized by vanishing gradients with an increasing number of qubits, hinder optimization in quantum algorithms. In this paper, we analyze the impact of state-of-the-art parameter initialization strategies from classical machine learning in random PQCs from the aspect of BP phenomenon. Our investigation encompasses a spectrum of initialization techniques, including random, Xavier (both normal and uniform variants), He, LeCun, and Orthogonal methods. Empirical assessment reveals a pronounced reduction in variance decay of gradients across all these methodologies compared to the randomly initialized PQCs. Specifically, the Xavier initialization technique outperforms the rest, showing a 62% improvement in variance decay compared to the random initialization. The He, Lecun, and orthogonal methods also display improvements, with respective enhancements of 32 %, 28 %, and 26 %. This compellingly suggests that the adoption of these existing initialization techniques holds the potential to significantly amplify the training efficacy of Quantum Neural Networks (QNNs), a subclass of PQCs. Demonstrating this effect, we employ the identified techniques to train QNNs for learning the identity function, effectively mitigating the adverse effects of BPs. The training performance, ranked from the best to the worst, aligns with the variance decay enhancement as outlined above. This paper underscores the role of tailored parameter initialization in mitigating the BP problem and eventually enhancing the training dynamics of QNNs.
This thesis addresses one of the central challenges in Quantum Machine Learning: the Barren Plateaus phenomenon, which causes the exponential vanishing of gradients and hin ders the effective training of variational quantum algorithms. We propose the Q-MOLLD method in which we design a hybrid protocol that leverages the generative capabilities of Large Language Models to generate quantum algorithms for optimization problems and to produce informed initializations for quantum circuits to improve their performance. Experimental results demonstrate that the proposed approach can mitigate the Barren Plateaus problem, preserving larger gradient magnitudes during the early stages of training and facilitating convergence compared to random or classical strategies. The findings suggest that Large Language Models can play a crucial role in the design and training of quantum algorithms, especially in the Noisy Intermediate-Scale Quantum era, where hardware limitations and noise pose significant challenges. Future work will focus on extending the analysis to different ansatz architectures, exploring active learning techniques, and evaluating performance under realistic noise conditions on quantum hardware.
Although recent advances in transfer learning have simplified training, fine-tuning big language models remains an energy-intensive and expensive process, with high hardware and energy costs that make it difficult to use and scale. While quantum machine learning (QML) presents a promising theoretical tool to overcome these bottlenecks, the practical implementation is hampered by the barren plateau problem, a phenomenon well established in literature that manifests through exponentially vanishing gradients of deep parameterized quantum circuits (PQCs) with increasing circuit depth, preventing informative parameter updates by rendering the optimization landscape flat. This study presents a hybrid quantum-classical architecture in which a PQC layer is integrated into a pretrained classical LLM backbone and fine-tuned throughout this work. This iterative refinement loop is controlled by an Expected Improvement acquisition function, which actively guides the search process through parameter regimes with obvious non-vanishing gradient variance. The study proposes that this theory-driven initialization scheme reduces the onset of barren plateaus, thereby enabling efficient and stable convergence in hybrid optimization. Experiments show that the proposed method consistently outperforms purely classical fine-tuning baselines and randomly initialized quantum baselines on representative downstream natural language processing tasks. The experimental results show that principled parameter initialization leads to concrete improvements in convergence stability, gradient trainability, and task-level performance. In summary, this research provides a scalable method for integrating near-term noisy intermediate-scale Quantum (NISQ) devices into state-of-the-art deep learning pipelines, fostering the further real-world adoption of hybrid quantum-classical systems for demanding artificial intelligence tasks.
Variational quantum algorithms often encounter barren plateaus, where cost gradients decay rapidly with increasing circuit depth, undermining the trainability of parameterized quantum circuits. This paper evaluates AdaInit (Adaptive Initialization), proposed by Zhuang and Cunningham, which uses large language models to propose initial parameters for quantum neural networks. We study a simplified single-query AdaInit variant paired with GPU-accelerated simulation in NVIDIA CUDA-Q and apply it to binary classification on the DMR-IR mammography dataset. AdaInit delivers 14.6 times higher gradient variance at initialization than random initialization (0.0095 vs. 0.0006), producing 160 times faster convergence (1.1s vs. 176 s) while maintaining the same classification accuracy of 61.4 percent. We provide theoretical analysis grounded in the geometry of parameterized circuit landscapes and show empirically that LLM-guided initialization places the optimizer in trainable regions of parameter space. Beyond performance, our results indicate that a single LLM query can yield informative parameters without iterative refinement, suggesting a low-overhead path to improved trainability. The findings validate AdaInit in a medical imaging setting and demonstrate its compatibility with GPU-accelerated quantum backends for practical speedups.
Barren plateaus (BPs) pose a major trainability challenge for quantum neural networks (QNNs) by causing gradients to concentrate near zero as circuit size, depth, or expressibility increases. This study presents a comparative benchmark of 10 BP mitigation strategies across six qubit settings (2, 4, 8, 12, 16, and 20) and three datasets of increasing complexity: Iris, MNIST, and MedMNIST. The evaluated methods include eight initialization-based strategies (Beta, Gaussian, Uniform Norm, CNN-based initialization, He-normal, He-uniform, Xavier-normal, and Xavier-uniform), one model-based variational encoder, and one optimization-based time-nonlocal Fourier parameterization. Experiments were implemented using PennyLane 3.10 and PyTorch 2.5 with simulator backends. We evaluate trainability using gradient variance and training loss, and we clarify that the benchmark analyzes simulated QNN optimization behavior rather than hardware-noise-resilient or noisy-label learning. Across the tested two-layer circuit configurations, the mitigation strategies maintained measurable gradient variance and stable loss reduction, suggesting that severe barren plateau behavior was not observed under the benchmark conditions. CNN-based and Beta initialization showed strong empirical behavior in variance retention and convergence speed, while Gaussian initialization was comparatively weaker in higher-dimensional settings. The study provides a reproducible benchmark structure for comparing BP mitigation behavior and identifies important limitations related to circuit depth, hardware noise, feature encoding, and classification performance that should be addressed in future QNN benchmarking.
Variational quantum algorithms (VQAs) are among the most promising algorithms in the era of noisy intermediate scale quantum devices. Such algorithms are constructed using a parametrization $U(\ensuremath{\theta}$) with a classical optimizer that updates the parameters $\ensuremath{\theta}$ in order to minimize a cost function C. For this task, in general the gradient descent method, or one of its variants, is used. This is a method where the circuit parameters are updated iteratively using the cost function gradient. However, several works in the literature have shown that this method suffers from a phenomenon known as the barren plateaus (BPs). In this paper, we propose a method to mitigate BPs. In general, the parameters $\ensuremath{\theta}$ used in the parametrization U are randomly generated. In our method they are obtained from a classical neural network (CNN). We show that this method, besides to being able to mitigate BPs during startup, is also able to mitigate the effect of BPs during the VQA training. In addition, we also show how this method behaves for different CNN architectures.