DeepAI AI Chat
Log In Sign Up

Prototypal Analysis and Prototypal Regression

by   Chenyue Wu, et al.
NYU college

Prototypal analysis is introduced to overcome two shortcomings of archetypal analysis: its sensitivity to outliers and its non-locality, which reduces its applicability as a learning tool. Same as archetypal analysis, prototypal analysis finds prototypes through convex combination of the data points and approximates the data through convex combination of the archetypes, but it adds a penalty for using prototypes distant from the data points for their reconstruction. Prototypal analysis can be extended---via kernel embedding---to probability distributions, since the convexity of the prototypes makes them interpretable as mixtures. Finally, prototypal regression is developed, a robust supervised procedure which allows the use of distributions as either features or labels.


page 1

page 2

page 3

page 4


On the repeated inversion of a covariance matrix

In many cases, the values of some model parameters are determined by max...

Twin Neural Network Regression is a Semi-Supervised Regression Algorithm

Twin neural network regression (TNNR) is a semi-supervised regression al...

Robust multivariate and functional archetypal analysis with application to financial time series analysis

Archetypal analysis approximates data by means of mixtures of actual ext...

Near-Convex Archetypal Analysis

Nonnegative matrix factorization (NMF) is a widely used linear dimension...

Survival Kernets: Scalable and Interpretable Deep Kernel Survival Analysis with an Accuracy Guarantee

Kernel survival analysis models estimate individual survival distributio...

Reduction of Two-Dimensional Data for Speeding Up Convex Hull Computation

An incremental approach for computation of convex hull for data points i...

Adaptive joint distribution learning

We develop a new framework for embedding (joint) probability distributio...