# Robust Mahalanobis Metric Learning via Geometric Approximation Algorithms

@article{Ihara2019RobustMM, title={Robust Mahalanobis Metric Learning via Geometric Approximation Algorithms}, author={Diego Ihara and Neshat Mohammadi and Francesco Sgherzi and Anastasios Sidiropoulos}, journal={arXiv: Learning}, year={2019} }

Learning Mahalanobis metric spaces is an important problem that has found numerous applications. Several algorithms have been designed for this problem, including Information Theoretic Metric Learning (ITML) [Davis et al. 2007] and Large Margin Nearest Neighbor (LMNN) classification [Weinberger and Saul 2009]. We study the problem of learning a Mahalanobis metric space in the presence of adversarial label noise. To that end, we consider a formulation of Mahalanobis metric learning as an…

## One Citation

Learning Lines with Ordinal Constraints

- Computer Science, MathematicsAPPROX-RANDOM
- 2020

This work studies the problem of finding a mapping from a set of points into the real line, under ordinal triple constraints, and presents an approximation algorithm for the dense case of this problem.

## References

SHOWING 1-10 OF 12 REFERENCES

Information-theoretic metric learning

- Computer ScienceICML '07
- 2007

An information-theoretic approach to learning a Mahalanobis distance function that can handle a wide variety of constraints and can optionally incorporate a prior on the distance function and derive regret bounds for the resulting algorithm.

Distance Metric Learning for Large Margin Nearest Neighbor Classification

- Computer ScienceNIPS
- 2005

This paper shows how to learn a Mahalanobis distance metric for kNN classification from labeled examples in a globally integrated manner and finds that metrics trained in this way lead to significant improvements in kNN Classification.

Metric Learning by Collapsing Classes

- Computer Science, MathematicsNIPS
- 2005

An algorithm for learning a quadratic Gaussian metric (Mahalanobis distance) for use in classification tasks and discusses how the learned metric may be used to obtain a compact low dimensional feature representation of the original input space, allowing more efficient classification with very little reduction in performance.

Metric Learning: A Survey

- Computer ScienceFound. Trends Mach. Learn.
- 2013

Metric Learning: A Review presents an overview of existing research in this topic, including recent progress on scaling to high-dimensional feature spaces and to data sets with an extremely large number of data points.

Distance Metric Learning with Application to Clustering with Side-Information

- Computer ScienceNIPS
- 2002

This paper presents an algorithm that, given examples of similar (and, if desired, dissimilar) pairs of points in �”n, learns a distance metric over ℝn that respects these relationships.

Geometric Approximation Algorithms

- Computer Science
- 2011

This book is the first to cover geometric approximation algorithms in detail, and topics covered include approximate nearest-neighbor search, shape approximation, coresets, dimension reduction, and embeddings.

Reconstructions of Noisy Digital Contours with Maximal Primitives Based on Multi-Scale/Irregular Geometric Representation and Generalized Linear Programming

- Computer Science
- 2017

A novel approach combining multi-scale and irregular isothetic representations of the input contour, as an extension of a previous work, improves the representation of the contour by 1-D intervals, and achieves afterwards the decomposition of thecontour into maximal arcs or segments.

The discrepancy method - randomness and complexity

- Computer Science
- 2000

This book tells the story of the discrepancy method in a few short independent vignettes. It is a varied tale which includes such topics as communication complexity, pseudo-randomness, rapidly mixing…

Linear programming and convex hulls made easy

- EconomicsSCG '90
- 1990

Two randomized algorithms that solve linear programs involving constraints in d variables in expected time and construct convex hulls of n points in Rsupscrpt (3) with dependence of the time bound on d.