Scalable and Incremental Learning of Gaussian Mixture Models
This work provides a scalable solution for incremental learning in high-dimensional data, which is incremental as it builds on existing methods with efficiency improvements.
The authors tackled the problem of efficiently learning Gaussian mixture models incrementally by introducing an algorithm that uses rank-one updates on precision matrices and determinants, achieving an asymptotic time complexity of O(NKD^2) for N data points, K components, and D dimensions, and demonstrated its applicability on classification datasets like MNIST and CIFAR-10 as well as reinforcement learning tasks.
This work presents a fast and scalable algorithm for incremental learning of Gaussian mixture models. By performing rank-one updates on its precision matrices and determinants, its asymptotic time complexity is of \BigO{NKD^2} for $N$ data points, $K$ Gaussian components and $D$ dimensions. The resulting algorithm can be applied to high dimensional tasks, and this is confirmed by applying it to the classification datasets MNIST and CIFAR-10. Additionally, in order to show the algorithm's applicability to function approximation and control tasks, it is applied to three reinforcement learning tasks and its data-efficiency is evaluated.