Robust Classification by Pre-conditioned LASSO and Transductive Diffusion Component Analysis

TL;DR

This paper introduces a robust classification framework that combines pre-conditioned LASSO for outlier detection with transductive diffusion component analysis to leverage data structure and improve accuracy in noisy datasets.

Contribution

It proposes a novel combination of pre-conditioned LASSO and transductive diffusion analysis for robust outlier detection and classification in large, noisy datasets.

Findings

Effective outlier detection in synthetic and real data

Improved classification accuracy with the proposed method

Robustness to noisy and unreliable datasets

Abstract

Modern machine learning-based recognition approaches require large-scale datasets with large number of labelled training images. However, such datasets are inherently difficult and costly to collect and annotate. Hence there is a great and growing interest in automatic dataset collection methods that can leverage the web. % which are collected % in a cheap, efficient and yet unreliable way. Collecting datasets in this way, however, requires robust and efficient ways for detecting and excluding outliers that are common and prevalent. % Outliers are thus a % prominent treat of using these dataset. So far, there have been a limited effort in machine learning community to directly detect outliers for robust classification. Inspired by the recent work on Pre-conditioned LASSO, this paper formulates the outlier detection task using Pre-conditioned LASSO and employs \red{unsupervised} transductive diffusion component analysis to both integrate the topological structure of the data manifold, from labeled and unlabeled instances, and reduce the feature dimensionality. Synthetic experiments as well as results on two real-world classification tasks show that our framework can robustly detect the outliers and improve classification.