Improving Pediatric Low-Grade Neuroepithelial Tumors Molecular Subtype Identification Using a Novel AUROC Loss Function for Convolutional Neural Networks.
Khashayar NamdarMatthias W. WagnerCynthia HawkinsUri TaboriBirgit B. Ertl-WagnerFarzad KhalvatiPublished in: CoRR (2024)
Keyphrases
- loss function
- convolutional neural networks
- pairwise
- support vector
- low grade
- logistic regression
- empirical risk
- learning to rank
- risk minimization
- convolutional network
- stochastic gradient descent
- special case
- boosting framework
- breast cancer
- boosting algorithms
- solution path
- image analysis
- convex loss functions
- feature selection