Exponential Families for Conditional Random Fields
This work addresses a theoretical and computational problem for researchers in machine learning, specifically in graphical models and kernel methods, but appears incremental as it builds on existing frameworks.
The paper tackles the problem of defining conditional random fields in reproducing kernel Hilbert spaces and connecting them to Gaussian Process classification, presenting efficient optimization methods using reduced rank decompositions and exploiting stationarity.
In this paper we de ne conditional random elds in reproducing kernel Hilbert spaces and show connections to Gaussian Process classi cation. More speci cally, we prove decomposition results for undirected graphical models and we give constructions for kernels. Finally we present e cient means of solving the optimization problem using reduced rank decompositions and we show how stationarity can be exploited e ciently in the optimization process.