Stochastic gradient updates yield deep equilibrium kernels.
Russell TsuchidaCheng Soon OngPublished in: Trans. Mach. Learn. Res. (2023)
Keyphrases
- stochastic gradient
- utility maximization
- stochastic gradient descent
- step size
- nearest neighbor classifier
- kernel function
- kernel methods
- grassmann manifold
- multiple kernel learning
- feature space
- support vector
- linear combination
- convergence rate
- utility function
- online learning
- worst case
- least squares
- evolutionary algorithm
- feature extraction