Using maximally entangled training data exponentially flattens the loss landscape of highly expressive quantum models, limiting the loss improvement achievable in a fixed-size neighborhood.
Universal compiling and (no-)free-lunch theorems for continuous-variable quantum learning.PRX Quantum, 2:040327, 2021
1 Pith paper cite this work, alongside 3 external citations. Polarity classification is still indexing.
1
Pith paper citing it
3
external citations · OpenAlex
fields
quant-ph 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Loss Behavior in Supervised Learning with Entangled States
Using maximally entangled training data exponentially flattens the loss landscape of highly expressive quantum models, limiting the loss improvement achievable in a fixed-size neighborhood.