A two-level Phase selection reformulation of real-valued phase retrieval, with simultaneous temperature and L2 regularization annealing, lets simulated annealing, AMP, and Langevin dynamics approach the Bayes-optimal sample efficiency in numerical experiments.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cond-mat.dis-nn 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Isolating the hard core of phaseless inference: the Phase selection formulation
A two-level Phase selection reformulation of real-valued phase retrieval, with simultaneous temperature and L2 regularization annealing, lets simulated annealing, AMP, and Langevin dynamics approach the Bayes-optimal sample efficiency in numerical experiments.