A fast, universal algorithm to learn parametric nonlinear embeddings.
Miguel Á. Carreira-PerpiñánMax VladymyrovPublished in: NIPS (2015)
Keyphrases
- experimental evaluation
- preprocessing
- times faster
- learning algorithm
- cost function
- convergence rate
- dynamic programming
- expectation maximization
- worst case
- computational cost
- computational complexity
- np hard
- optimal solution
- computationally efficient
- theoretical analysis
- image sequences
- detection algorithm
- piecewise linear
- recognition algorithm
- improved algorithm
- neural network
- convex hull
- segmentation algorithm
- linear programming
- probabilistic model
- significant improvement
- k means
- artificial neural networks