Reproduction Capabilities of Penalized Hyperbolic-polynomial Splines

02/14/2022
by   Rosanna Campagna, et al.
0

This paper investigates two important analytical properties of hyperbolic-polynomial penalized splines, HP-splines for short. HP-splines, obtained by combining a special type of difference penalty with hyperbolic-polynomial B-splines (HB-splines), were recently introduced by the authors as a generalization of P-splines. HB-splines are bell-shaped basis functions consisting of segments made of real exponentials e^α x, e^-α x and linear functions multiplied by these exponentials, xe^+α x and xe^-α x. Here, we show that these type of penalized splines reproduce function in the space {e^-α x, x e^-α x}, that is they fit exponential data exactly. Moreover, we show that they conserve the first and second 'exponential' moments.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/15/2020

A Theory of Hyperbolic Prototype Learning

We introduce Hyperbolic Prototype Learning, a type of supervised learnin...
research
12/17/2020

Annihilation Operators for Exponential Spaces in Subdivision

We investigate properties of differential and difference operators annih...
research
02/11/2020

A sufficient condition for penalized polynomial regression to be invariant to translations of the predictor variables

Whereas translating the coding of predictor variables does not change th...
research
05/21/2021

Generalization Error Bound for Hyperbolic Ordinal Embedding

Hyperbolic ordinal embedding (HOE) represents entities as points in hype...
research
02/24/2020

A Quantized Representation of Intertemporal Choice in the Brain

Value [4][5] is typically modeled using a continuous representation (i.e...
research
06/07/2013

Non-constant bounded holomorphic functions of hyperbolic numbers - Candidates for hyperbolic activation functions

The Liouville theorem states that bounded holomorphic complex functions ...
research
06/07/2022

Towards Scalable Hyperbolic Neural Networks using Taylor Series Approximations

Hyperbolic networks have shown prominent improvements over their Euclide...

Please sign up or login with your details

Forgot password? Click here to reset