Distance metric learning based on different ℓ1 regularized metric matrices in each category.
Kenta MikawaManabu KobayashiMasayuki GotoShigeichi HirasawaPublished in: ISITA (2016)
Keyphrases
- distance metric learning
- metric learning
- distance metric
- nearest neighbor classification
- semi supervised
- learning tasks
- pairwise
- euclidean distance
- distance function
- distance measure
- image classification
- singular value decomposition
- multi task
- dimensionality reduction
- feature space
- least squares
- data points
- similarity measure
- background knowledge
- covariance matrices
- machine learning algorithms
- high dimensional
- object categories
- gender classification
- object recognition
- objective function