Simulated polariton condensate lattices act as physics-based feature generators for CNNs and improve classification of cliques and asymmetries in point clouds over raw point images in three synthetic tasks.
Excitonic oscillator-strength saturation dominates polariton-polariton interactions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Exciton-polaritons in semiconductor microcavities exhibit large two-body interactions that, thanks to ever refined nanotechnology techniques, are getting closer and closer to the quantum regime where single-photon nonlinearities start being relevant. To foster additional progress in this direction, in this work we experimentally investigate the microscopic mechanism driving polariton-polariton interactions. We measure the dispersion relation of the collective excitations that are thermally generated on top of a coherent fluid of interacting lower-polaritons. By comparing the measurements with the Bogoliubov theory over both the lower and upper polariton branches simultaneously, we find that polariton-polariton interactions stem dominantly from a mechanism of saturation of the exciton oscillator strength.
citation-role summary
background 1
citation-polarity summary
fields
cond-mat.dis-nn 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Polaritonic Machine Learning for Graph-based Data Analysis
Simulated polariton condensate lattices act as physics-based feature generators for CNNs and improve classification of cliques and asymmetries in point clouds over raw point images in three synthetic tasks.