Whole Slide Pathology Image Patch Based Deep Classification: An Investigation of the Effects of the Latent Autoencoder Representation and the Loss Function Form.
Ana LomacenkovaOgnjen ArandjelovicPublished in: BHI (2021)
Keyphrases
- loss function
- image classification
- image representation
- support vector
- image features
- hinge loss
- high resolution
- machine learning
- image retrieval
- image segmentation
- support vector machine
- classification accuracy
- feature vectors
- pairwise
- bayes rule
- image patches
- test images
- texture synthesis
- training set
- learning to rank
- text classification
- unsupervised learning
- empirical risk
- convex loss functions
- logistic regression
- model selection
- low dimensional
- supervised learning
- multi class
- similarity measure
- decision trees
- feature selection