DeepAI AI Chat
Log In Sign Up

Learning Data Manifolds with a Cutting Plane Method

by   SueYeon Chung, et al.
Hebrew University of Jerusalem
University of Pennsylvania
Harvard University

We consider the problem of classifying data manifolds where each manifold represents invariances that are parameterized by continuous degrees of freedom. Conventional data augmentation methods rely upon sampling large numbers of training examples from these manifolds; instead, we propose an iterative algorithm called M_CP based upon a cutting-plane approach that efficiently solves a quadratic semi-infinite programming problem to find the maximum margin solution. We provide a proof of convergence as well as a polynomial bound on the number of iterations required for a desired tolerance in the objective function. The efficiency and performance of M_CP are demonstrated in high-dimensional simulations and on image manifolds generated from the ImageNet dataset. Our results indicate that M_CP is able to rapidly learn good classifiers and shows superior generalization performance compared with conventional maximum margin methods using data augmentation methods.


page 1

page 2

page 3

page 4


Isometric Multi-Manifolds Learning

Isometric feature mapping (Isomap) is a promising manifold learning meth...

Stochastic Cutting Planes for Data-Driven Optimization

We introduce a stochastic version of the cutting-plane method for a larg...

Cardinality-constrained Distributionally Robust Portfolio Optimization

This paper studies a distributionally robust portfolio optimization mode...

When does data augmentation help generalization in NLP?

Neural models often exploit superficial ("weak") features to achieve goo...

Using Self-Organising Mappings to Learn the Structure of Data Manifolds

In this paper it is shown how to map a data manifold into a simpler form...