An Elementary Approach to Convergence Guarantees of Optimization Algorithms for Deep Networks.
Vincent RouletZaïd HarchaouiPublished in: Allerton (2019)
Keyphrases
- data sets
- optimization problems
- global convergence
- discrete optimization
- convergence rate
- learning algorithm
- theoretical justification
- optimization methods
- optimization algorithm
- theoretical analysis
- computational cost
- data streams
- data structure
- evolutionary algorithm
- orders of magnitude
- reinforcement learning
- graph theory
- real world networks
- iterative algorithms
- bayesian networks
- decision trees
- stochastic gradient
- clustering algorithm
- quasi newton
- convergence proof