Kernel Conditional Density Operators
This provides a kernel-based method for conditional density estimation, which is incremental relative to existing approaches.
The authors tackled the problem of conditional density estimation by introducing the conditional density operator (CDO), which captures multivariate, multimodal output densities and achieves competitive performance with neural models and Gaussian processes on synthetic and real-world data.
We introduce a novel conditional density estimation model termed the conditional density operator (CDO). It naturally captures multivariate, multimodal output densities and shows performance that is competitive with recent neural conditional density models and Gaussian processes. The proposed model is based on a novel approach to the reconstruction of probability densities from their kernel mean embeddings by drawing connections to estimation of Radon-Nikodym derivatives in the reproducing kernel Hilbert space (RKHS). We prove finite sample bounds for the estimation error in a standard density reconstruction scenario, independent of problem dimensionality. Interestingly, when a kernel is used that is also a probability density, the CDO allows us to both evaluate and sample the output density efficiently. We demonstrate the versatility and performance of the proposed model on both synthetic and real-world data.