Metric learning via perturbing hard-to-classify instances.
Xinyao GuoWei WeiJianqing LiangChuangyin DangJiye LiangPublished in: Pattern Recognit. (2022)
Keyphrases
- metric learning
- distance metric
- distance metric learning
- semi supervised
- pairwise
- machine learning and pattern recognition
- dimensionality reduction
- multi task
- learning tasks
- person re identification
- semi supervised clustering
- distance function
- semi supervised learning
- nearest neighbor classification
- mahalanobis metric
- semi definite programming
- feature space
- subject to linear constraints
- maximum variance unfolding
- neural network
- labeled data
- distance measure
- multi class
- knowledge base
- data mining