Abstract:Quantum 8, 1484 (2024). https://doi.org/10.22331/q-2024-09-25-1484 The training of a parameterized model largely depends on the landscape of the underlying loss function. In particular, vanishing gradients are a central bottleneck in the scalability of variational quantum algorithms (VQAs), and are known to arise in various ways. However, a caveat of most existing gradient bound results is the requirement of t-design circuit assumptions that are typically not satisfied in practice. In this work, we loosen these assumptions altogether and derive tight upper and lower bounds on loss and gradient concentration for a large class of parameterized quantum circuits and arbitrary observables, which are significantly stronger than prior work. Moreover, we show that these bounds, as well as the variance of the loss itself, can be estimated efficiently and classically-providing practical tools to study the loss landscapes of VQA models, including verifying whether or not a circuit/observable induces barren plateaus. In particular, our results can readily be leveraged to rule out barren plateaus for a realistic class of ansätze and mixed observables, namely, observables containing a non-vanishing local term. This insight has direct implications for hybrid Quantum Generative Adversarial Networks (qGANs). We prove that designing the discriminator appropriately leads to 1-local weights that stay constant in the number of qubits, regardless of discriminator depth. This implies that qGANs with appropriately chosen generators do not suffer from barren plateaus even at scale-making them a promising candidate for applications in generative quantum machine learning. We demonstrate this result by training a qGAN to learn a 2D mixture of Gaussian distributions with up to 16 qubits, and provide numerical evidence that global contributions to the gradient, while initially exponentially small, may kick in substantially over the course of training.

Toward Trainability of Deep Quantum Neural Networks

Toward Trainability of Quantum Neural Networks

Absence of Barren Plateaus in Quantum Convolutional Neural Networks

Optimizing Quantized Neural Networks in a Weak Curvature Manifold

Statistical Analysis of Quantum State Learning Process in Quantum Neural Networks

Quantum Generative Training Using Rényi Divergences

Trade-off between Gradient Measurement Efficiency and Expressivity in Deep Quantum Neural Networks

A quantum algorithm for training wide and deep classical neural networks

Escaping from the Barren Plateau Via Gaussian Initializations in Deep Variational Quantum Circuits

The Dilemma of Quantum Neural Networks

Deep quantum neural networks form Gaussian processes

A Unified Theory of Quantum Neural Network Loss Landscapes

Tight and Efficient Gradient Bounds for Parameterized Quantum Circuits

Learning To Optimize Quantum Neural Network Without Gradients

Dynamical transition in controllable quantum neural networks with large depth

Analyzing Convergence in Quantum Neural Networks: Deviations from Neural Tangent Kernels

Infinite Neural Network Quantum States: Entanglement and Training Dynamics

Training-efficient density quantum machine learning

Training Classical Neural Networks by Quantum Machine Learning

Enhanced Universality in Quantum Neural Networks with Fewer Qubits and Reduced Circuit Depth