اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدAccelerated Stochastic ADMM with Variance Reduction
67
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Alternating Direction Method of Multipliers (ADMM) is a popular method in solving Machine Learning problems. Stochastic ADMM was firstly proposed in order to reduce the per iteration computational complexity, which is more suitable for big data problems. Recently, variance reduction techniques have been integrated with stochastic ADMM in order to get a fast convergence rate, such as SAG-ADMM and SVRG-ADMM,but the convergence is still suboptimal w.r.t the smoothness constant. In this paper, we propose a new accelerated stochastic ADMM algorithm with variance reduction, which enjoys a faster convergence than all the other stochastic ADMM algorithms. We theoretically analyze its convergence rate and show its dependence on the smoothness constant is optimal. We also empirically validate its effectiveness and show its priority over other stochastic ADMM algorithms.
قيم البحث
اقرأ أيضاً
Stochastic gradient Langevin dynamics (SGLD) has gained the attention of optimization researchers due to its global optimization properties. This paper proves an improved convergence property to local minimizers of nonconvex objective functions using
SGLD accelerated by variance reductions. Moreover, we prove an ergodicity property of the SGLD scheme, which gives insights on its potential to find global minimizers of nonconvex objectives.
In this paper, we develop a symmetric accelerated stochastic Alternating Direction Method of Multipliers (SAS-ADMM) for solving separable convex optimization problems with linear constraints. The objective function is the sum of a possibly nonsmooth
convex function and an average function of many smooth convex functions. Our proposed algorithm combines both ideas of ADMM and the techniques of accelerated stochastic gradient methods using variance reduction to solve the smooth subproblem. One main feature of SAS-ADMM {is} that its dual variable is symmetrically updated after each update of the separated primal variable, which would allow a more flexible and larger convergence region of the dual variable compared with that of standard deterministic or stochastic ADMM. This new stochastic optimization algorithm is shown to converge in expectation with $C{O}(1/T)$ convergence rate, where $T$ is the number of outer iterations. In addition, 3-block extensions of the algorithm and its variant of an accelerated stochastic augmented Lagrangian method are also discussed. Our preliminary numerical experiments indicate the proposed algorithm is very effective for solving separable optimization problems from big-data applications
The variance reduction class of algorithms including the representative ones, SVRG and SARAH, have well documented merits for empirical risk minimization problems. However, they require grid search to tune parameters (step size and the number of iter
ations per inner loop) for optimal performance. This work introduces `almost tune-free SVRG and SARAH schemes equipped with i) Barzilai-Borwein (BB) step sizes; ii) averaging; and, iii) the inner loop length adjusted to the BB step sizes. In particular, SVRG, SARAH, and their BB variants are first reexamined through an `estimate sequence lens to enable new averaging methods that tighten their convergence rates theoretically, and improve their performance empirically when the step size or the inner loop length is chosen large. Then a simple yet effective means to adjust the number of iterations per inner loop is developed to enhance the merits of the proposed averaging schemes and BB step sizes. Numerical tests corroborate the proposed methods.
Variance reduction (VR) methods for finite-sum minimization typically require the knowledge of problem-dependent constants that are often unknown and difficult to estimate. To address this, we use ideas from adaptive gradient methods to propose AdaSV
RG, which is a fully adaptive variant of SVRG, a common VR method. AdaSVRG uses AdaGrad in the inner loop of SVRG, making it robust to the choice of step-size, and allowing it to adaptively determine the length of each inner-loop. When minimizing a sum of $n$ smooth convex functions, we prove that AdaSVRG requires $O(n + 1/epsilon)$ gradient evaluations to achieve an $epsilon$-suboptimality, matching the typical rate, but without needing to know problem-dependent constants. However, VR methods including AdaSVRG are slower than SGD when used with over-parameterized models capable of interpolating the training data. Hence, we also propose a hybrid algorithm that can adaptively switch from AdaGrad to AdaSVRG, achieving the best of both stochastic gradient and VR methods, but without needing to tune their step-sizes. Via experiments on synthetic and standard real-world datasets, we validate the robustness and effectiveness of AdaSVRG, demonstrating its superior performance over other tune-free VR methods.
An inexact accelerated stochastic Alternating Direction Method of Multipliers (AS-ADMM) scheme is developed for solving structured separable convex optimization problems with linear constraints. The objective function is the sum of a possibly nonsmoo
th convex function and a smooth function which is an average of many component convex functions. Problems having this structure often arise in machine learning and data mining applications. AS-ADMM combines the ideas of both ADMM and the stochastic gradient methods using variance reduction techniques. One of the ADMM subproblems employs a linearization technique while a similar linearization could be introduced for the other subproblem. For a specified choice of the algorithm parameters, it is shown that the objective error and the constraint violation are $mathcal{O}(1/k)$ relative to the number of outer iterations $k$. Under a strong convexity assumption, the expected iterate error converges to zero linearly. A linearized variant of AS-ADMM and incremental sampling strategies are also discussed. Numerical experiments with both stochastic and deterministic ADMM algorithms show that AS-ADMM can be particularly effective for structured optimization arising in big data applications.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
