Second‐Order Parabolic Differential Equations

Adam and RMSProp are two of the most influential adaptive stochastic algorithms for training deep neural networks, which have been pointed out to be divergent even in the convex setting via a few simple counterexamples. Many attempts, such as decreasing an adaptive learning rate, adopting a big batch size, incorporating a temporal decorrelation technique, seeking an analogous surrogate, etc., have been tried to promote Adam/RMSProp-type algorithms to converge. In contrast with existing approaches, we introduce an alternative easy-to-check sufficient condition, which merely depends on the parameters of the base learning rate and combinations of historical second-order moments, to guarantee the global convergence of generic Adam/RMSProp for solving large-scale non-convex stochastic optimization. Moreover, we show that the convergences of several variants of Adam, such as AdamNC, AdaEMA, etc., can be directly implied via the proposed sufficient condition in the non-convex setting. In addition, we illustrate that Adam is essentially a specifically weighted AdaGrad with exponential moving average momentum, which provides a novel perspective for understanding Adam and RMSProp. This observation coupled with this sufficient condition gives much deeper interpretations on their divergences. At last, we validate the sufficient condition by applying Adam and RMSProp to tackle a certain counterexample and train deep neural networks. Numerical results are exactly in accord with our theoretical analysis.

/pdf/a-sufficient-condition-for-convergences-of-adam-and-rmsprop-457gekuo6m.pdf

A Sufficient Condition for Convergences of Adam and RMSProp

Scaling up the convolutional neural network (CNN) size (e.g., width, depth, etc.) is known to effectively improve model accuracy. However, the large model size impedes training on resource-constrained edge devices. For instance, federated learning (FL) may place undue burden on the compute capability of edge nodes, even though there is a strong practical need for FL due to its privacy and confidentiality properties. To address the resource-constrained reality of edge devices, we reformulate FL as a group knowledge transfer training algorithm, called FedGKT. FedGKT designs a variant of the alternating minimization approach to train small CNNs on edge nodes and periodically transfer their knowledge by knowledge distillation to a large server-side CNN. FedGKT consolidates several advantages into a single framework: reduced demand for edge computation, lower communication bandwidth for large CNNs, and asynchronous training, all while maintaining model accuracy comparable to FedAvg. We train CNNs designed based on ResNet-56 and ResNet-110 using three distinct datasets (CIFAR-10, CIFAR-100, and CINIC-10) and their non-I.I.D. variants. Our results show that FedGKT can obtain comparable or even slightly higher accuracy than FedAvg. More importantly, FedGKT makes edge training affordable. Compared to the edge training using FedAvg, FedGKT demands 9 to 17 times less computational power (FLOPs) on edge devices and requires 54 to 105 times fewer parameters in the edge CNN. Our source code is released at FedML (this https URL).

/pdf/group-knowledge-transfer-federated-learning-of-large-cnns-at-1zecnktrrv.pdf

Group Knowledge Transfer: Federated Learning of Large CNNs at the Edge

This paper studies a class of adaptive gradient based momentum algorithms that update the search directions and learning rates simultaneously using past gradients. This class, which we refer to as the "Adam-type", includes the popular algorithms such as the Adam, AMSGrad and AdaGrad. Despite their popularity in training deep neural networks, the convergence of these algorithms for solving nonconvex problems remains an open question. This paper provides a set of mild sufficient conditions that guarantee the convergence for the Adam-type methods. We prove that under our derived conditions, these methods can achieve the convergence rate of order $O(\log{T}/\sqrt{T})$ for nonconvex stochastic optimization. We show the conditions are essential in the sense that violating them may make the algorithm diverge. Moreover, we propose and analyze a class of (deterministic) incremental adaptive gradient algorithms, which has the same $O(\log{T}/\sqrt{T})$ convergence rate. Our study could also be extended to a broader class of adaptive gradient methods in machine learning and optimization.

/pdf/on-the-convergence-of-a-class-of-adam-type-algorithms-for-p6i2t4e1nw.pdf

On the convergence of a class of Adam-type algorithms for non-convex optimization

Adaptive Moment Estimation (Adam) is a very popular training algorithm for deep neural networks, implemented in many machine learning frameworks. To the best of the authors knowledge no complete convergence analysis exists for Adam. The contribution of this paper is a method for the local convergence analysis in batch mode for a deterministic fixed training set, which gives necessary conditions for the hyperparameters of the Adam algorithm. Due to the local nature of the arguments the objective function can be non-convex but must be at least twice continuously differentiable.

A Proof of Local Convergence for the Adam Optimizer

A Sufficient Condition for Convergences of Adam and RMSProp.

Integrating adaptive learning rate and momentum techniques into SGD leads to a large class of efficiently accelerated adaptive stochastic algorithms, such as Nadam, AccAdaGrad, \textit{etc}. In spite of their effectiveness in practice, there is still a large gap in their theories of convergences, especially in the difficult non-convex stochastic setting. To fill this gap, we propose \emph{weighted AdaGrad with unified momentum}, dubbed AdaUSM, which has the main characteristics that (1) it incorporates a unified momentum scheme which covers both the heavy ball momentum and the Nesterov accelerated gradient momentum; (2) it adopts a novel weighted adaptive learning rate that can unify the learning rates of AdaGrad, AccAdaGrad, Adam, and RMSProp. Moreover, when we take polynomially growing weights in AdaUSM, we obtain its $\mathcal{O}(\log(T)/\sqrt{T})$ convergence rate in the non-convex stochastic setting. We also show that the adaptive learning rates of Adam and RMSProp correspond to taking exponentially growing weights in AdaUSM, which thereby provides a new perspesctive for understanding Adam and RMSProp. Lastly, comparative experiments of AdaUSM against SGD with momentum, AdaGrad, AdaEMA, Adam, and AMSGrad on various deep learning models and datasets are also provided.

Weighted AdaGrad with Unified Momentum

On the Convergence of AdaGrad with Momentum for Training Deep Neural Networks

Adaptive stochastic gradient descent methods, such as AdaGrad, RMSProp, Adam, AMSGrad, etc., have been demonstrated efficacious in solving non-convex stochastic optimization, such as training deep neural networks. However, their convergence rates have not been touched under the non-convex stochastic circumstance except recent breakthrough results on AdaGrad, perturbed AdaGrad and AMSGrad. In this paper, we propose two new adaptive stochastic gradient methods called AdaHB and AdaNAG which integrate a novel weighted coordinate-wise AdaGrad with heavy ball momentum and Nesterov accelerated gradient momentum, respectively. The $\mathcal{O}(\frac{\log{T}}{\sqrt{T}})$ non-asymptotic convergence rates of AdaHB and AdaNAG in non-convex stochastic setting are also jointly established by leveraging a newly developed unified formulation of these two momentum mechanisms. Moreover, comparisons have been made between AdaHB, AdaNAG, Adam and RMSProp, which, to a certain extent, explains the reasons why Adam and RMSProp are divergent. In particular, when momentum term vanishes we obtain convergence rate of coordinate-wise AdaGrad in non-convex stochastic setting as a byproduct.

Fangyu Zou

Papers

A Sufficient Condition for Convergences of Adam and RMSProp

A Sufficient Condition for Convergences of Adam and RMSProp.

Weighted AdaGrad with Unified Momentum

On the Convergence of AdaGrad with Momentum for Training Deep Neural Networks

On the Convergence of Weighted AdaGrad with Momentum for Training Deep Neural Networks