اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدAccelerating Gossip SGD with Periodic Global Averaging
232
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Communication overhead hinders the scalability of large-scale distributed training. Gossip SGD, where each node averages only with its neighbors, is more communication-efficient than the prevalent parallel SGD. However, its convergence rate is reversely proportional to quantity $1-beta$ which measures the network connectivity. On large and sparse networks where $1-beta to 0$, Gossip SGD requires more iterations to converge, which offsets against its communication benefit. This paper introduces Gossip-PGA, which adds Periodic Global Averaging into Gossip SGD. Its transient stage, i.e., the iterations required to reach asymptotic linear speedup stage, improves from $Omega(beta^4 n^3/(1-beta)^4)$ to $Omega(beta^4 n^3 H^4)$ for non-convex problems. The influence of network topology in Gossip-PGA can be controlled by the averaging period $H$. Its transient-stage complexity is also superior to Local SGD which has order $Omega(n^3 H^4)$. Empirical results of large-scale training on image classification (ResNet50) and language modeling (BERT) validate our theoretical findings.
قيم البحث
اقرأ أيضاً
Federated Learning (FL) is a recent development in the field of machine learning that collaboratively trains models without the training data leaving client devices, to preserve data privacy. In realistic FL settings, the training set is distributed
over clients in a highly non-Independent and Identically Distributed (non-IID) fashion, which has been shown extensively to harm FL convergence speed and final model performance. To address this challenge, we propose a novel, generalised approach for incorporating adaptive optimisation techniques into FL with the Federated Global Biased Optimiser (FedGBO) algorithm. FedGBO accelerates FL by employing a set of global biased optimiser values during the client-training phase, which helps to reduce `client-drift from non-IID data, whilst also benefiting from adaptive optimisation. We show that the FedGBO update with a generic optimiser can be reformulated as centralised training using biased gradients and optimiser updates, and apply this theoretical framework to prove the convergence of FedGBO using momentum-Stochastic Gradient Descent (SGDm). We also conduct extensive experiments using 4 realistic FL benchmark datasets (CIFAR100, Sent140, FEMNIST, Shakespeare) and 3 popular adaptive optimisers (RMSProp, SGDm, Adam) to compare the performance of state-of-the-art adaptive-FL algorithms. The results demonstrate that FedGBO has highly competitive performance whilst achieving lower communication and computation costs, and provide practical insights into the trade-offs associated with the different adaptive-FL algorithms and optimisers for real-world FL deployments.
Nesterov SGD is widely used for training modern neural networks and other machine learning models. Yet, its advantages over SGD have not been theoretically clarified. Indeed, as we show in our paper, both theoretically and empirically, Nesterov SGD w
ith any parameter selection does not in general provide acceleration over ordinary SGD. Furthermore, Nesterov SGD may diverge for step sizes that ensure convergence of ordinary SGD. This is in contrast to the classical results in the deterministic scenario, where the same step size ensures accelerated convergence of the Nesterovs method over optimal gradient descent. To address the non-acceleration issue, we introduce a compensation term to Nesterov SGD. The resulting algorithm, which we call MaSS, converges for same step sizes as SGD. We prove that MaSS obtains an accelerated convergence rates over SGD for any mini-batch size in the linear setting. For full batch, the convergence rate of MaSS matches the well-known accelerated rate of the Nesterovs method. We also analyze the practically important question of the dependence of the convergence rate and optimal hyper-parameters on the mini-batch size, demonstrating three distinct regimes: linear scaling, diminishing returns and saturation. Experimental evaluation of MaSS for several standard architectures of deep networks, including ResNet and convolutional networks, shows improved performance over SGD, Nesterov SGD and Adam.
Large-scale distributed training of neural networks is often limited by network bandwidth, wherein the communication time overwhelms the local computation time. Motivated by the success of sketching methods in sub-linear/streaming algorithms, we intr
oduce Sketched SGD, an algorithm for carrying out distributed SGD by communicating sketches instead of full gradients. We show that Sketched SGD has favorable convergence rates on several classes of functions. When considering all communication -- both of gradients and of updated model weights -- Sketched SGD reduces the amount of communication required compared to other gradient compression methods from $mathcal{O}(d)$ or $mathcal{O}(W)$ to $mathcal{O}(log d)$, where $d$ is the number of model parameters and $W$ is the number of workers participating in training. We run experiments on a transformer model, an LSTM, and a residual network, demonstrating up to a 40x reduction in total communication cost with no loss in final model performance. We also show experimentally that Sketched SGD scales to at least 256 workers without increasing communication cost or degrading model performance.
Stochastic Gradient Descent (SGD) is the key learning algorithm for many machine learning tasks. Because of its computational costs, there is a growing interest in accelerating SGD on HPC resources like GPU clusters. However, the performance of paral
lel SGD is still bottlenecked by the high communication costs even with a fast connection among the machines. A simple approach to alleviating this problem, used in many existing efforts, is to perform communication every few iterations, using a constant averaging period. In this paper, we show that the optimal averaging period in terms of convergence and communication cost is not a constant, but instead varies over the course of the execution. Specifically, we observe that reducing the variance of model parameters among the computing nodes is critical to the convergence of periodic parameter averaging SGD. Given a fixed communication budget, we show that it is more beneficial to synchronize more frequently in early iterations to reduce the initial large variance and synchronize less frequently in the later phase of the training process. We propose a practical algorithm, named ADaptive Periodic parameter averaging SGD (ADPSGD), to achieve a smaller overall variance of model parameters, and thus better convergence compared with the Constant Periodic parameter averaging SGD (CPSGD). We evaluate our method with several image classification benchmarks and show that our ADPSGD indeed achieves smaller training losses and higher test accuracies with smaller communication compared with CPSGD. Compared with gradient-quantization SGD, we show that our algorithm achieves faster convergence with only half of the communication. Compared with full-communication SGD, our ADPSGD achieves 1:14x to 1:27x speedups with a 100Gbps connection among computing nodes, and the speedups increase to 1:46x ~ 1:95x with a 10Gbps connection.
Stochastic gradient descent (SGD) has been widely studied in the literature from different angles, and is commonly employed for solving many big data machine learning problems. However, the averaging technique, which combines all iterative solutions
into a single solution, is still under-explored. While some increasingly weighted averaging schemes have been considered in the literature, existing works are mostly restricted to strongly convex objective functions and the convergence of optimization error. It remains unclear how these averaging schemes affect the convergence of {it both optimization error and generalization error} (two equally important components of testing error) for {bf non-strongly convex objectives, including non-convex problems}. In this paper, we {it fill the gap} by comprehensively analyzing the increasingly weighted averaging on convex, strongly convex and non-convex objective functions in terms of both optimization error and generalization error. In particular, we analyze a family of increasingly weighted averaging, where the weight for the solution at iteration $t$ is proportional to $t^{alpha}$ ($alpha > 0$). We show how $alpha$ affects the optimization error and the generalization error, and exhibit the trade-off caused by $alpha$. Experiments have demonstrated this trade-off and the effectiveness of polynomially increased weighted averaging compared with other averaging schemes for a wide range of problems including deep learning.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
