On the capacity and superposition of minima in neural network loss function landscapes.
Maximilian P. NiroomandJohn W. R. MorganConor T. CafollaDavid J. WalesPublished in: Mach. Learn. Sci. Technol. (2022)
Keyphrases
- loss function
- neural network
- pairwise
- logistic regression
- learning to rank
- support vector
- back propagation
- empirical risk
- regularization term
- risk minimization
- reproducing kernel hilbert space
- machine learning
- hinge loss
- boosting framework
- stochastic gradient descent
- training algorithm
- solution path
- ensemble methods
- multi class