A Deep Learning Architecture for Egocentric Time-to-Saccade Prediction using Weibull Mixture-Models and Historic Priors.
Tim RolffSusanne SchmidtFrank SteinickeSimone FrintropPublished in: ETRA (2023)
Keyphrases
- mixture model
- deep learning
- unsupervised learning
- gaussian mixture model
- dirichlet prior
- probabilistic model
- em algorithm
- generative model
- expectation maximization
- supervised learning
- mixture modeling
- language model
- maximum likelihood
- mental models
- machine learning
- feature selection
- computer vision
- dimensionality reduction
- image processing
- weakly supervised
- data mining
- model selection
- higher order
- eye movements