Improving Unsupervised Domain Adaptive Re-Identification via Source-Guided Selection of Pseudo-Labeling Hyperparameters.
Fabian DubourvieuxAngélique LoeschRomaric AudigierSamia AinouzStéphane CanuPublished in: CoRR (2021)
Keyphrases
- hyperparameters
- model selection
- cross validation
- closed form
- bayesian inference
- maximum likelihood
- random sampling
- em algorithm
- bayesian framework
- gaussian process
- active learning
- support vector
- gaussian processes
- semi supervised
- maximum a posteriori
- noise level
- grid search
- parameter space
- worst case
- support vector machine
- image segmentation
- genetic algorithm