Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userGeneralized Framework for Nonlinear Acceleration
75
0
0.0
(
0
)
Authors
Damien Scieur
Ask ChatGPT about the research
No Arabic abstract
Nonlinear acceleration algorithms improve the performance of iterative methods, such as gradient descent, using the information contained in past iterates. However, their efficiency is still not entirely understood even in the quadratic case. In this paper, we clarify the convergence analysis by giving general properties that share several classes of nonlinear acceleration: Anderson acceleration (and variants), quasi-Newton methods (such as Broyden Type-I or Type-II, SR1, DFP, and BFGS) and Krylov methods (Conjugate Gradient, MINRES, GMRES). In particular, we propose a generic family of algorithms that contains all the previous methods and prove its optimal rate of convergence when minimizing quadratic functions. We also propose multi-secants updates for the quasi-Newton methods listed above. We provide a Matlab code implementing the algorithm.
rate research
Read More
The Regularized Nonlinear Acceleration (RNA) algorithm is an acceleration method capable of improving the rate of convergence of many optimization schemes such as gradient descend, SAGA or SVRG. Until now, its analysis is limited to convex problems, but empirical observations shows that RNA may be extended to wider settings. In this paper, we investigate further the benefits of RNA when applied to neural networks, in particular for the task of image recognition on CIFAR10 and ImageNet. With very few modifications of exiting frameworks, RNA improves slightly the optimization process of CNNs, after training.
In this paper, we introduce various mechanisms to obtain accelerated first-order stochastic optimization algorithms when the objective function is convex or strongly convex. Specifically, we extend the Catalyst approach originally designed for deterministic objectives to the stochastic setting. Given an optimization method with mild convergence guarantees for strongly convex problems, the challenge is to accelerate convergence to a noise-dominated region, and then achieve convergence with an optimal worst-case complexity depending on the noise variance of the gradients. A side contribution of our work is also a generic analysis that can handle inexact proximal operators, providing new insights about the robustness of stochastic algorithms when the proximal operator cannot be exactly computed.
This work describes a family of attitude estimators that are based on a generalization of Mahonys nonlinear complementary filter. This generalization reveals the close mathematical relationship between the nonlinear complementary filter and the more traditional multiplicative extended Kalman filter. In fact, the bias-free and constant gain multiplicative continuous-time extended Kalman filters may be interpreted as special cases of the generalized attitude estimator. The correspondence provides a rational means of choosing the gains for the nonlinear complementary filter and a proof of the near global asymptotic stability of special cases of the multiplicative extended Kalman filter.
We describe convergence acceleration schemes for multistep optimization algorithms. The extrapolated solution is written as a nonlinear average of the iterates produced by the original optimization method. Our analysis does not need the underlying fixed-point operator to be symmetric, hence handles e.g. algorithms with momentum terms such as Nesterovs accelerated method, or primal-dual methods. The weights are computed via a simple linear system and we analyze performance in both online and offline modes. We use Crouzeixs conjecture to show that acceleration performance is controlled by the solution of a Chebyshev problem on the numerical range of a non-symmetric operator modeling the behavior of iterates near the optimum. Numerical experiments are detailed on logistic regression problems.
The most important ingredient for solving mixed-integer nonlinear programs (MINLPs) to global epsilon-optimality with spatial branch and bound is a tight, computationally tractable relaxation. Due to both theoretical and practical considerations, relaxations of MINLPs are usually required to be convex. Nonetheless, current optimization solver can often successfully handle a moderate presence of nonconvexities, which opens the door for the use of potentially tighter nonconvex relaxations. In this work, we exploit this fact and make use of a nonconvex relaxation obtained via aggregation of constraints: a surrogate relaxation. These relaxations were actively studied for linear integer programs in the 70s and 80s, but they have been scarcely considered since. We revisit these relaxations in an MINLP setting and show the computational benefits and challenges they can have. Additionally, we study a generalization of such relaxation that allows for multiple aggregations simultaneously and present the first algorithm that is capable of computing the best set of aggregations. We propose a multitude of computational enhancements for improving its practical performance and evaluate the algorithms ability to generate strong dual bounds through extensive computational experiments.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
