Adjoint shadowing for backpropagation in hyperbolic chaos

07/14/2022
by   Angxiu Ni, et al.
0

For both discrete-time and continuous-time hyperbolic chaos, we introduce the adjoint shadowing operator 𝒮 acting on covector fields. We show that 𝒮 can be equivalently defined as: (a) 𝒮 is the adjoint of the linear shadowing operator; (b) 𝒮 is given by a `split then propagate' expansion formula; (c) 𝒮(ω) is the only bounded inhomogeneous adjoint solution of ω. By (a), 𝒮 adjointly expresses the shadowing contribution, the most significant part of the linear response, where the linear response is the derivative of the long-time statistics with respect to parameters. By (b), 𝒮 also expresses the other part of the linear response, the unstable contribution. By (c), 𝒮 can be efficiently computed by the nonintrusive shadowing algorithm, which is similar to the conventional backpropagation algorithm. For continuous-time cases, we additionally show that the linear response admits a well-defined decomposition into shadowing and unstable contributions.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset
Success!
Error Icon An error occurred

Sign in with Google

×

Use your Google Account to sign in to DeepAI

×

Consider DeepAI Pro