nonconvex optimization
Convergence of Cubic Regularization for Nonconvex Optimization under KL Property
Yi Zhou, Zhe Wang, Yingbin Liang
Cubic-regularized Newton's method (CR) is a popular algorithm that guarantees
- North America > United States > Ohio (0.04)
- North America > Canada > Quebec > Montreal (0.04)
- Asia > Middle East > Jordan (0.04)
A Stochastic Composite Gradient Method with Incremental Variance Reduction
Neural Information Processing Systems http://nips.cc/
- North America > United States > Minnesota > Hennepin County > Minneapolis (0.28)
- Oceania > Australia > New South Wales > Sydney (0.04)
- North America > United States > New York > New York County > New York City (0.04)
- (5 more...)
Stochastic Nested Variance Reduced Gradient Descent for Nonconvex Optimization
Dongruo Zhou, Pan Xu, Quanquan Gu
Neural Information Processing Systems http://nips.cc/
- North America > United States > California > Los Angeles County > Los Angeles (0.29)
- North America > Canada > Quebec > Montreal (0.04)
- Europe > Russia (0.04)
- (2 more...)
ZO-AdaMM: Zeroth-Order Adaptive Momentum Method for Black-Box Optimization
Xiangyi Chen, Sijia Liu, Kaidi Xu, Xingguo Li, Xue Lin, Mingyi Hong, David Cox
Neural Information Processing Systems http://nips.cc/
- North America > United States > Minnesota (0.04)
- North America > Canada (0.04)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.04)
- (2 more...)
- Asia > Middle East > Jordan (0.04)
- North America > United States > California (0.04)
- Europe > United Kingdom > England > Cambridgeshire > Cambridge (0.04)
Gradient-FreeMethodsforDeterministicand StochasticNonsmoothNonconvexOptimization
Thus a gap has arisen between theory and practice.
- Asia > Middle East > Jordan (0.05)
- North America > Canada > Ontario > Toronto (0.04)
- North America > United States > Washington > King County > Seattle (0.04)
- North America > United States > Massachusetts > Middlesex County > Cambridge (0.04)
- Asia > Middle East > Jordan (0.04)
Zeroth-Order Methods for Nondifferentiable, Nonconvex, and Hierarchical Federated Optimization
Federated learning (FL) has emerged as an enabling framework for communication-efficient decentralized training.
- North America > United States > Pennsylvania (0.04)
- Asia > Middle East > Jordan (0.04)
An Inexact Augmented Lagrangian Framework for Nonconvex Optimization with Nonlinear Constraints
We propose a practical inexact augmented Lagrangian method (iALM) for nonconvex problems with nonlinear constraints. We characterize the total computational complexity of our method subject to a verifiable geometric condition, which is closely related to the Polyak-Lojasiewicz and Mangasarian-Fromowitz conditions. In particular, when a first-order solver is used for the inner iterates, we prove that iALM finds a first-order stationary point with $\tilde{\mathcal{O}}(1/\epsilon^3)$ calls to the first-order oracle.