Improves quantum sample complexity for learning Fourier-sparse Boolean functions from O(k^1.5 log²k) to Õ(k) via symmetric differences and multilinear majority polynomials, achieving near-optimal bounds and closing a decade-long theoretical gap.

#QuantumLearning #QuantumAlgorithms #Research

Weak Schur sampling remains robust to independent per-copy noise, with output error bounded by noise floor ε plus O(d/√n)—proving the foundational Keyl-Werner algorithm for spectrum estimation survives realistic noisy quantum devices.

#QuantumLearning #QuantumTomography #Research

Derives universal generalization bounds for learning quantum unitaries across different input distributions, showing real-state training controls complex-state performance exponentially well.

#QuantumLearning #QuantumAlgorithms #Research

Resolves optimal sample complexity for estimating quantum state moments Tr(ρᵗ) with adaptive single-copy measurements, revealing a dyadic hierarchy where orders 3 and 4 both require Θ(d²/³) copies despite differing nonadaptively.

#QuantumLearning #QuantumMeasurement #Research

2nd AIMS Workshop on the Theory of Quantum Learning Algorithms
📅 30 Nov – 3 Dec 2026
📍 AIMS, Muizenberg
🎓 Funding is available for up to 20 students & postdoctoral researchers at SA universities.
buff.ly/1W9yqvY
#QuantumLearning #QuantumComputing #LearningAlgorithms #AIMSSouthAfrica #QuantumResearch

Breakthrough quantum learning algorithm: efficiently learns low-bond-dimension quantum states from data without assuming the unknown state belongs to the model class, using compression and dynamic programming with polynomial complexity.

#QuantumLearning #TensorNetworks #Research

New quantum score matching framework extends classical learning technique to quantum states, achieving optimal sample complexity for high-temperature Gibbs states. NISQ-friendly implementation on IBM hardware reduces parameter error from 64% to 10%.

#QuantumLearning #QuantumAlgorithms #Research