Continuous-time Value Function Approximation in Reproducing Kernel Hilbert Spaces.
Motoya OhnishiMasahiro YukawaMikael JohanssonMasashi SugiyamaPublished in: NeurIPS (2018)
Keyphrases
- reproducing kernel hilbert space
- state space
- reproducing kernel
- loss function
- uniform convergence
- learning theory
- gaussian kernels
- kernel methods
- basis functions
- euclidean space
- covering numbers
- kernel function
- density estimation
- input space
- special case
- data dependent
- gaussian process
- distance measure
- domain adaptation
- representer theorem
- real valued
- kernel matrix
- learning problems
- reinforcement learning
- linear combination
- stochastic processes
- hilbert space
- euclidean distance
- pairwise
- training set