SEVIN verifies a neural network steering controller on low-dimensional latent polytopes built from a GM-VAE, claiming equivalence to verification on real images.
Fundamental limits to signal integrity in nonlinear parametric optical circulators
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We characterize the response of a parametric nonlinear optical circulator to realistic signals that have finite bandwidths. Our results show that intermodulation distortion (IMD), rather than pump depletion or compression, limits the maximal operating signal power and the dynamic range of nonlinear parametric circulators. This limitation holds even in the undepleted pump regime where nonlinear circulators are not constrained by dynamic reciprocity. With a realistic pump power, noise floor, and nonlinear waveguide, our numerical modeling demonstrates a maximally achievable spur-free dynamic range (SFDR) of 81 dB.
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Scalable and Interpretable Verification of Image-based Neural Network Controllers for Autonomous Vehicles
SEVIN verifies a neural network steering controller on low-dimensional latent polytopes built from a GM-VAE, claiming equivalence to verification on real images.