Proofs and additional experiments on Second order techniques for learning time-series with structural breaks

12/15/2020

∙

We provide complete proofs of the lemmas about the properties of the regularized loss function that is used in the second order techniques for learning time-series with structural breaks in Osogami (2021). In addition, we show experimental results that support the validity of the techniques.

READ FULL TEXT

Proofs and additional experiments on Second order techniques for learning time-series with structural breaks

Modelling Time-Varying First and Second-Order Structure of Time Series via Wavelets and Differencing

Global Constraint Catalog, Volume II, Time-Series Constraints

Detecting structural breaks in eigensystems of functional time series

Structural time series grammar over variable blocks

Batch Normalization Is Blind to the First and Second Derivatives of the Loss

Several Coercivity Proofs of First-Order System Least-Squares Methods for Second-Order Elliptic PDEs

On a Proof of the Convergence Speed of a Second-order Recurrence Formula in the Arimoto-Blahut Algorithm

Proofs and additional experiments on Second order techniques for learning time-series with structural breaks

Related Research

Modelling Time-Varying First and Second-Order Structure of Time Series via Wavelets and Differencing

Global Constraint Catalog, Volume II, Time-Series Constraints

Detecting structural breaks in eigensystems of functional time series

Structural time series grammar over variable blocks

Batch Normalization Is Blind to the First and Second Derivatives of the Loss

Several Coercivity Proofs of First-Order System Least-Squares Methods for Second-Order Elliptic PDEs

On a Proof of the Convergence Speed of a Second-order Recurrence Formula in the Arimoto-Blahut Algorithm