Nonlinear Kalman Filtering with Reparametrization Gradients
This work addresses a specific bottleneck in state-space modeling for researchers in signal processing or machine learning, but it appears incremental as it builds on existing techniques like alpha divergence and reparametrization gradients.
The paper tackles the problem of nonlinear Kalman filtering by introducing a new method that uses reparametrization gradients to optimize an energy function, resulting in improved approximations compared to traditional Gaussian-based approaches.
We introduce a novel nonlinear Kalman filter that utilizes reparametrization gradients. The widely used parametric approximation is based on a jointly Gaussian assumption of the state-space model, which is in turn equivalent to minimizing an approximation to the Kullback-Leibler divergence. It is possible to obtain better approximations using the alpha divergence, but the resulting problem is substantially more complex. In this paper, we introduce an alternate formulation based on an energy function, which can be optimized instead of the alpha divergence. The optimization can be carried out using reparametrization gradients, a technique that has recently been utilized in a number of deep learning models.