Barren plateaus are swamped with traps

Abstract

Two main challenges preventing efficient training of variational quantum algorithms and quantum machine learning models are local minima and barren plateaus. Typically, barren plateaus are associated with deep circuits, while shallow circuits have been shown to suffer from suboptimal local minima. We point out a simple mechanism that creates exponentially many poor local minima specifically in the barren plateau regime. These local minima are trivial solutions, optimizing only a few terms in the loss function, leaving the rest on their barren plateaus. More precisely, we show the existence of approximate local minima, optimizing a single loss term, and conjecture the existence of exact local minima, optimizing only a logarithmic fraction of all loss function terms. One implication of our findings is that simply yielding large gradients is not sufficient to render an initialization strategy a meaningful solution to the barren plateau problem.

9+5 pages

References in corpus (14)

• An introduction to quantum machine learning
• Beyond Barren Plateaus: Quantum Variational Algorithms Are Swamped With Traps
• Exploiting symmetry in variational quantum machine learning
• Group-Invariant Quantum Machine Learning
• A Lie Algebraic Theory of Barren Plateaus for Deep Parameterized Quantum Circuits
• The Adjoint Is All You Need: Characterizing Barren Plateaus in Quantum Ansätze
• Synergy Between Quantum Circuits and Tensor Networks: Short-cutting the Race to Practical Quantum Advantage
• Classical surrogates for quantum learning models
• Training Saturation in Layerwise Quantum Approximate Optimisation
• Trainability Enhancement of Parameterized Quantum Circuits via Reduced-Domain Parameter Initialization
• Multidimensional Fourier series with quantum circuits
• Fourier expansion in variational quantum algorithms
• Efficient variational synthesis of quantum circuits with coherent multi-start optimization
• Equivalence of cost concentration and gradient vanishing for quantum circuits: An elementary proof in the Riemannian formulation

Cited by in corpus (4)

• Barren Plateaus in Variational Quantum Computing
• Deep-Circuit QAOA
• Evidence for effectively constant shot complexity in the quantum approximate optimization algorithm without per-instance optimization
• Learning complexity gradually in quantum machine learning models