H-Likelihood Approach to Deep Neural Networks with Temporal-Spatial Random Effects for High-Cardinality Categorical Features.
Hangbin LeeYoungjo LeePublished in: ICML (2023)
Keyphrases
- neural network
- spatial features
- spatial and temporal
- spatio temporal
- random effects
- spatial information
- temporal dimension
- temporal information
- space time
- temporal correlation
- image features
- artificial neural networks
- maximum likelihood
- feature vectors
- spatial relationships
- feature space
- temporal relationships
- feature extraction
- back propagation
- feature set
- temporal aspects
- computer vision