Presentation
High-order Differentiable Autoencoder for Nonlinear Model Reduction
SessionTechnical Papers
Event Type
Technical Paper
Research & Education
Ultimate Supporter
Ultimate Attendee
Exhibitor Ultimate
Time
Location
DescriptionCombing BP, we show that complex-step finite difference can be used for computing high-order differentiation of a deep net. This new method allows a new autoencoder-based nonlinear model reduction.