Trainability ยท Algorithms

Orthogonal circuits exhibit a barren plateau

Theorem statement

For the circuit family whose dynamical Lie algebra is $\mathfrak{so}(2^n)$, the single-ideal loss variance $$\operatorname{Var}_\theta[\ell]=\frac{P_{\mathfrak g}(\rho)\,P_{\mathfrak g}(O)}{\dim\mathfrak g},\qquad \dim\mathfrak g=\frac{4^n-2^n}{2},$$ carries an algebra dimension that grows exponentially in the qubit count $n$. With the purities $P_{\mathfrak g}(\rho),\,P_{\mathfrak g}(O)$ bounded, the variance decays as $C\,b^{-n}$ with $b>1$, so the family exhibits a barren plateau.

Sources

  1. A Lie Algebraic Theory of Barren Plateaus for Deep Parameterized Quantum Circuits

    Michael Ragone, Bojko N. Bakalov, Frederic Sauvage, Alexander F. Kemper, Carlos Ortiz Marrero, Martin Larocca, M. Cerezo, 2023

  2. Barren plateaus in quantum neural network training landscapes

    Jarrod R. McClean, Sergio Boixo, Vadim N. Smelyanskiy, Ryan Babbush, Hartmut Neven, 2018

Lean context

Copy a short prompt with the import, theorem name, citations, and public source link.

Open Lean source