Neural Bayesian Filtering
This addresses state estimation in partially observable environments, offering a hybrid approach that is incremental in combining classical filters with deep generative models.
The paper tackles the problem of maintaining distributions over hidden states in partially observable systems by introducing Neural Bayesian Filtering (NBF), which maps beliefs to embedding vectors for efficient particle-style updates, and validates it in state estimation tasks across three environments.
We present Neural Bayesian Filtering (NBF), an algorithm for maintaining distributions over hidden states, called beliefs, in partially observable systems. NBF is trained to find a good latent representation of the beliefs induced by a task. It maps beliefs to fixed-length embedding vectors, which condition generative models for sampling. During filtering, particle-style updates compute posteriors in this embedding space using incoming observations and the environment's dynamics. NBF combines the computational efficiency of classical filters with the expressiveness of deep generative models - tracking rapidly shifting, multimodal beliefs while mitigating the risk of particle impoverishment. We validate NBF in state estimation tasks in three partially observable environments.