Similarity learning in nearest neighbor, positive semi-definitiveness and RELIEF algorithm.
Ali Mustafa QamarÉric GaussierPublished in: SoCPaR (2010)
Keyphrases
- learning algorithm
- nearest neighbor
- learning speed
- k means
- detection algorithm
- computational complexity
- significant improvement
- nearest neighborhood
- clustering method
- dynamic programming
- distance metric
- matching algorithm
- distance function
- cost function
- similarity measure
- probabilistic model
- learning process
- simulated annealing
- online learning
- expectation maximization
- incremental learning
- positive and negative
- learning tasks
- learned models
- knn
- data sets
- optimal solution
- objective function
- reinforcement learning
- image segmentation
- neural network