Pitfalls when tackling the exponential concentration of parameterized quantum models
arXiv:2507.22054 · doi:10.1088/2058-9565/ae2202
Abstract
Identifying scalable circuit architectures remains a central challenge in variational quantum computing and quantum machine learning. Many approaches have been proposed to mitigate or avoid the barren plateau phenomenon or, more broadly, exponential concentration. However, due to the intricate interplay between quantum measurements and classical post-processing, we argue these techniques often fail to circumvent concentration effects in practice. Here, by analyzing concentration at the level of measurement outcome probabilities and leveraging tools from hypothesis testing, we develop a practical framework for diagnosing whether a parameterized quantum model is inhibited by exponential concentration. Applying this framework, we argue that several widely used methods (including quantum natural gradient, sample-based optimization, and certain neural-network-inspired initializations) do not overcome exponential concentration with finite measurement budgets, though they may still aid training in other ways.
11+13 pages, 5 figures
References in corpus (56)
- Variational Quantum Algorithms
- Supervised learning with quantum enhanced feature spaces
- Barren plateaus in quantum neural network training landscapes
- The theory of variational hybrid quantum-classical algorithms
- Noisy intermediate-scale quantum (NISQ) algorithms
- Quantum computational chemistry
- Quantum Circuit Learning
- Predicting Many Properties of a Quantum System from Very Few Measurements
- Cost Function Dependent Barren Plateaus in Shallow Parametrized Quantum Circuits
- Evaluating analytic gradients on quantum hardware
- Power of data in quantum machine learning
- Data re-uploading for a universal quantum classifier
- Connecting ansatz expressibility to gradient magnitudes and barren plateaus
- Quantum advantage in learning from experiments
- Quantum Natural Gradient
- An initialization strategy for addressing barren plateaus in parametrized quantum circuits
- The randomized measurement toolbox
- Quantum-assisted quantum compiling
- Harnessing disordered quantum dynamics for machine learning
- Effect of barren plateaus on gradient-free optimization
- Improving Variational Quantum Optimization using CVaR
- Beyond Barren Plateaus: Quantum Variational Algorithms Are Swamped With Traps
- Diagnosing Barren Plateaus with Tools from Quantum Optimal Control
- Barren Plateaus in Variational Quantum Computing
- Trainability of Dissipative Perceptron-Based Quantum Neural Networks
- Theory of overparametrization in quantum neural networks
- Opportunities in Quantum Reservoir Computing and Extreme Learning Machines
- Group-Invariant Quantum Machine Learning
- Avoiding local minima in variational quantum eigensolvers with the natural gradient optimizer
- Equivalence of quantum barren plateaus to cost concentration and narrow gorges
- A Lie Algebraic Theory of Barren Plateaus for Deep Parameterized Quantum Circuits
- Entanglement Devised Barren Plateau Mitigation
- Boosting computational power through spatial multiplexing in quantum reservoir computing
- Barren plateaus preclude learning scramblers
- Higher Order Derivatives of Quantum Neural Networks with Barren Plateaus
- Theoretical Guarantees for Permutation-Equivariant Quantum Neural Networks
- Barren plateaus in quantum tensor network optimization
- Hamiltonian variational ansatz without barren plateaus
- Avoiding barren plateaus via transferability of smooth solutions in Hamiltonian Variational Ansatz
- The Adjoint Is All You Need: Characterizing Barren Plateaus in Quantum Ansätze
- Reconstructing quantum states with quantum reservoir networks
- Benchmarking of Different Optimizers in the Variational Quantum Algorithms for Applications in Quantum Chemistry
- Classical surrogates for quantum learning models
- Expressibility and trainability of parameterized analog quantum systems for machine learning applications
- Trainability barriers and opportunities in quantum generative modeling
- Absence of barren plateaus in finite local-depth circuits with long-range entanglement
- Trainability Enhancement of Parameterized Quantum Circuits via Reduced-Domain Parameter Initialization
- Quantum Deep Hedging
- Variational quantum simulation: a case study for understanding warm starts
- Tight and Efficient Gradient Bounds for Parameterized Quantum Circuits
- Tackling Sampling Noise in Physical Systems for Machine Learning Applications: Fundamental Limits and Eigentasks
- Neural network encoded variational quantum algorithms
- Dequantizing quantum machine learning models using tensor networks
- Using Differential Evolution to avoid local minima in Variational Quantum Algorithms
- Potential and limitations of random Fourier features for dequantizing quantum machine learning
- Optimized numerical gradient and Hessian estimation for variational quantum algorithms