Deep Metric Learning Using Negative Sampling Probability Annealing.
Gábor KertészPublished in: Sensors (2022)
Keyphrases
- metric learning
- distance metric
- distance metric learning
- semi supervised
- learning tasks
- pairwise
- machine learning and pattern recognition
- dimensionality reduction
- kernel matrix
- multi task
- semi supervised clustering
- distance function
- mahalanobis metric
- nearest neighbor classification
- feature space
- semi supervised learning
- sample size
- simulated annealing
- subject to linear constraints
- random sampling
- nearest neighbor
- euclidean distance
- learning experience
- active learning
- maximum variance unfolding
- query processing
- learning algorithm