Toward Joint Optimization of Circuit Depth and Training Data Size in Adaptively Grown Quantum Classifiers

Key points

• Q-FLAIR shows that a quantum feature-map circuit can be grown gate-by-gate, stopping once further growth stops improving the training loss.1
• Does Q-FLAIR's own stopping rule pick larger or smaller circuits as training data grows?1
• We reimplement Q-FLAIR's growth mechanism faithfully, including its analytic reconstruction and exact stopping rule.1
• We find no predictable relationship between training-set size and the circuit size Q-FLAIR converges to.1

Sources (1)

• [1]Toward Joint Optimization of Circuit Depth and Training Data Size in Adaptively Grown Quantum Classifiers
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 05:55 PM

  Building a quantum model involves a tradeoff: how complex the circuit should be, and how much training data it needs.

  Q-FLAIR shows that a quantum feature-map circuit can be grown gate-by-gate, stopping once further growth stops improving the training loss.

Extractive summary: sentences quoted from the sources.