Toward Joint Optimization of Circuit Depth and Training Data Size in Adaptively Grown Quantum Classifiers
What it is
This paper investigates whether the adaptive circuit growth mechanism of Q-FLAIR and the generalization bounds of Caro et al. can be combined to predict optimal quantum model scaling. It empirically tests these relationships on MNIST classification, finding that while the theoretical bound holds, it has weak predictive power for actual circuit size and accuracy trends.
The problem it solves
The lack of a predictable scaling law that jointly optimizes quantum circuit complexity and training data requirements, leading to uncertainty in model design.
What you could use it for
- Guiding the design of adaptive quantum machine learning algorithms
- Analyzing the relationship between circuit complexity and dataset size in quantum classifiers
- Validating theoretical generalization bounds in practical quantum training scenarios
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.