SAFE ma-QAOA achieves 64.3% fewer active parameters and 94.5% lower estimated QPU workload via surrogate pre-training and parameter distillation on Sherrington-Kirkpatrick, 2D spin glass, and Max-Cut instances.
The Quantum Approximate Optimization Algorithm and the Sherrington-Kirkpatrick Model at Infinite Size.Quantum, 6: 759, July 2022
8 Pith papers cite this work, alongside 166 external citations. Polarity classification is still indexing.
8
Pith papers citing it
166
external citations · external index
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
CRiSP uses neural-guided MCTS and curriculum learning to insert Clifford prefixes before parameterized rotations in VQAs, yielding mean 3.17x and max 45x gains in energy accuracy on 22-qubit QAOA benchmarks versus prior Clifford initializers.
Zero-noise extrapolation has a finite-shot help-harm boundary below which it increases local mean-squared error due to variance penalties outweighing bias reduction.
Analytical reweighting rules for QAOA parameters on hypergraphs improve performance by adjusting mixing terms beyond previous graph-based methods.
Global Bradley-Terry rankings of LLMs are misleading due to structured heterogeneity in user preferences, and small (λ, ν)-portfolios recover coherent subpopulations that cover over 96% of votes with just five rankings.
Large qLDPC blocks in distributed quantum computing enable Pauli-based computation to run up to 10x faster than surface codes for optimization algorithms by using spare nodes to bypass serialization bottlenecks.
Systematic numerical study of QAOA parameter transfer on heavy-hex Ising models with local cubic terms shows transferred angles from small instances yield improving expectation values up to 49 layers on instances up to 156 qubits, with hardware runs confirming gains up to p=10.
TETRIS-ADAPT-VQE achieves fidelities above 99.3% for SYK (N=20) and 99.9998% for SK (L=18) but requires large resources for SYK models.
citing papers explorer
-
SAFE ma-QAOA: Surrogate-Assisted and Fine-Tuning Enhanced Multi-Angle QAOA with Parameter Distillation
SAFE ma-QAOA achieves 64.3% fewer active parameters and 94.5% lower estimated QPU workload via surrogate pre-training and parameter distillation on Sherrington-Kirkpatrick, 2D spin glass, and Max-Cut instances.
-
Classical State Preparation for Variational Quantum Algorithms via Reinforcement Learning
CRiSP uses neural-guided MCTS and curriculum learning to insert Clifford prefixes before parameterized rotations in VQAs, yielding mean 3.17x and max 45x gains in energy accuracy on 22-qubit QAOA benchmarks versus prior Clifford initializers.
-
The finite-shot help-harm boundary of zero-noise extrapolation
Zero-noise extrapolation has a finite-shot help-harm boundary below which it increases local mean-squared error due to variance penalties outweighing bias reduction.
-
QAOA Parameter Transfer for Hypergraphs
Analytical reweighting rules for QAOA parameters on hypergraphs improve performance by adjusting mixing terms beyond previous graph-based methods.
-
Why Global LLM Leaderboards Are Misleading: Small Portfolios for Heterogeneous Supervised ML
Global Bradley-Terry rankings of LLMs are misleading due to structured heterogeneity in user preferences, and small (λ, ν)-portfolios recover coherent subpopulations that cover over 96% of votes with just five rankings.
-
Space-Time Tradeoffs of Pauli-Based Computation in Distributed qLDPC Architectures
Large qLDPC blocks in distributed quantum computing enable Pauli-based computation to run up to 10x faster than surface codes for optimization algorithms by using spare nodes to bypass serialization bottlenecks.
-
Evaluating the Limits of QAOA Parameter Transfer at High-Rounds on Sparse Ising Models With Geometrically Local Cubic Terms
Systematic numerical study of QAOA parameter transfer on heavy-hex Ising models with local cubic terms shows transferred angles from small instances yield improving expectation values up to 49 layers on instances up to 156 qubits, with hardware runs confirming gains up to p=10.
-
Ground state preparation of random all-to-all Hamiltonians using ADAPT-VQE
TETRIS-ADAPT-VQE achieves fidelities above 99.3% for SYK (N=20) and 99.9998% for SK (L=18) but requires large resources for SYK models.