Business Intelligence

Some New Directions in Graph-Based Semi-Supervised Learning

Free registration required

Executive Summary

In this position paper, the authors first review the state-of-the-art in graph-based semi-supervised learning, and point out three limitations that are particularly relevant to multimedia analysis: Rich data is restricted to live on a single manifold; learning must happen in batch mode; and the target label is assumed smooth on the manifold. They then discuss new directions in semi-supervised learning research that can potentially overcome these limitations: modeling data as a mixture of multiple manifolds that may intersect or overlap; online semi-supervised learning that learns incrementally with low computation and memory needs; and learning spectrally sparse but non-smooth labels with compressive sensing. They give concrete examples in each new direction.

  • Format: PDF
  • Size: 791.36 KB