Kernel Conditional Density Operators
arXiv:1905.11255
Abstract
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.
References in corpus (11)
- GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration
- MMD GAN: Towards Deeper Understanding of Moment Matching Network
- Generative Moment Matching Networks
- Kernel Interpolation for Scalable Structured Gaussian Processes (KISS-GP)
- Training generative neural networks via Maximum Mean Discrepancy optimization
- Density Estimation in Infinite Dimensional Exponential Families
- Large sample analysis of the median heuristic
- Scaling Gaussian Process Regression with Derivatives
- Scalable Gaussian Processes with Grid-Structured Eigenfunctions (GP-GRIEF)
- Blitzkriging: Kronecker-structured Stochastic Gaussian Processes
- Noise Contrastive Meta-Learning for Conditional Density Estimation using Kernel Mean Embeddings