A Non-isotropic Probabilistic Take on Proxy-based Deep Metric Learning.
Michael KirchhofKarsten RothZeynep AkataEnkelejda KasneciPublished in: CoRR (2022)
Keyphrases
- metric learning
- distance metric
- distance metric learning
- semi supervised
- pairwise
- nearest neighbor classification
- learning tasks
- multi task
- dimensionality reduction
- distance function
- machine learning and pattern recognition
- semi supervised clustering
- subject to linear constraints
- person re identification
- probabilistic model
- feature space
- mahalanobis metric
- semi supervised learning
- multi class
- training data
- kernel learning
- active learning
- image segmentation
- semi definite programming
- image processing
- maximum variance unfolding
- learning algorithm