The game theoretic <i>p</i> -Laplacian and semi-supervised learning with few labels

Jeff Calder

Nonlinearity · 2018 · 59 citations · 32 references

Concepts

Abstract

Abstract We study the game theoretic p -Laplacian for semi-supervised learning on graphs, and show that it is well-posed in the limit of finite labeled data and infinite unlabeled data. In particular, we show that the continuum limit of graph-based semi-supervised learning with the game theoretic p -Laplacian is a weighted version of the continuous p -Laplace equation. We also prove that solutions to the graph p -Laplace equation are approximately Hölder continuous with high probability. Our proof uses the viscosity solution machinery and the maximum principle on a graph.

References

32