STORM+: Fully Adaptive SGD with Recursive Momentum for Nonconvex Optimization.
Kfir Y. LevyAli KavisVolkan CevherPublished in: NeurIPS (2021)
Keyphrases
- optimization problems
- global optimization
- nonlinear programming
- short term
- stochastic gradient descent
- stochastic gradient
- long term
- learning rate
- combinatorial optimization
- subgradient method
- recursive algorithm
- convex optimization
- optimal solution
- optimization method
- optimization algorithm
- cost function
- lower bound
- feature space