HCNAF: Hyper-Conditioned Neural Autoregressive Flow and its Application for Probabilistic Occupancy Map Forecasting

Geunseob Oh, Jean-Sebastien Valois
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We introduce Hyper-Conditioned Neural Autoregressive Flow (HCNAF); a powerful universal distribution approximator designed to model arbitrarily complex conditional probability density functions. HCNAF consists of a neural-net based conditional autoregressive flow (AF) and a hyper-network that can take large conditions in nonautoregressive fashion and outputs the network parameters of the AF. Like other flow models, HCNAF performs exact likelihood inference. We conduct a number of density
more » ... ion tasks on toy experiments and MNIST to demonstrate the effectiveness and attributes of HCNAF, including its generalization capability over unseen conditions and expressivity. Finally, we show that HCNAF scales up to complex high-dimensional prediction problems of the magnitude of self-driving and that HCNAF yields a state-of-theart performance in a public self-driving dataset.
doi:10.1109/cvpr42600.2020.01456 dblp:conf/cvpr/OhV20 fatcat:7eyqfncr6jgczaldaebmzzgovu