Mitigating Barren Plateaus in Quantum Neural Networks via an AI-Driven Submartingale-Based Framework
This addresses a critical bottleneck in training quantum neural networks for NISQ computing applications, though it is an incremental improvement over initialization-based strategies.
The paper tackles the problem of barren plateaus in quantum neural networks by proposing AdaInit, a framework that uses generative models with submartingale properties to synthesize initial parameters, resulting in consistently higher gradient variance across different scales compared to existing methods.
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 in terms of the qubit size. Most existing initialization-based mitigation strategies rely heavily on pre-designed 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 generative models with the submartingale property to iteratively synthesize initial parameters for QNNs that yield non-negligible 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.