Informed Spectral Normalized Gaussian Processes for Trajectory Prediction

arXiv:2403.11966v1 Announce Type: new
Abstract: Prior parameter distributions provide an elegant way to represent prior expert and world knowledge for informed learning. Previous work has shown that using such informative priors to regularize probabilistic deep learning (DL) models increases their performance and data-efficiency. However, commonly used sampling-based approximations for probabilistic DL models can be computationally expensive, requiring multiple inference passes and longer training times. Promising alternatives are compute-efficient last layer kernel approximations like spectral normalized Gaussian processes (SNGPs). We propose …

abstract arxiv cs.ai cs.lg data deep learning efficiency expert gaussian processes however knowledge performance prediction prior probabilistic deep learning processes sampling trajectory type work world