Ph.D. student Fanghui Liu's paper "A Double-Variational Bayesian Framework in Random Fourier Features for Indefinite Kernels" (by Fanghui Liu, Xiaolin Huang, Lei Shi, Jie Yang, and Johan A.K. Suykens) was accepted by IEEE Transactions on Neural Networks and Learning Systems.  

     This paper proposes a double-infinite Gaussian mixture model in RFF by placing the Dirichlet process prior. It takes full advantage of high flexibility on the number of components and has the capability of approximating indefinite kernels on a wide scale. In model inference, this paper develops a non-conjugate variational algorithm with a sub-sampling scheme for posterior inference. It allows for the non-conjugate case in the model and is quite efficient due to the sub-sampling strategy.