UD-SGD analysis shows efficient sampling by a few agents can outperform others.
problem Analyzing convergence speed and sampling strategies in UD-SGD.
method Asymptotic analysis of UD-SGD with various communication patterns and sampling strategies.
result Efficient sampling by a few agents can lead to better overall convergence.
Unified analysis of SGD variants for nonconvex federated optimization.
problem Performance of stochastic gradient methods in nonconvex optimization.
method Proposed a unified assumption for modeling stochastic gradient second moment, leading to a single convergence analysis for various methods.
result Unified convergence analysis for a wide range of SGD variants and distributed methods.
Unified analysis of asynchronous-SGD algorithms for distributed learning.
problem Analyzing asynchronous-SGD in heterogeneous settings with varying speeds and data distributions.
method Unified convergence theory for non-convex smooth functions, including pure asynchronous SGD and its modifications.
result Unified convergence rates for various asynchronous algorithms, including novel methods.
Communication-efficient SGD algorithms, which allow nodes to perform local updates and periodically synchronize local models, are highly effective in improving the speed and scalability of distributed SGD. However, a rigorous convergence analysis and comparative study of different communication-reduction strategies rem…
Unified analysis for decentralized SGD across various topologies and updates.
problem Analysis of decentralized SGD methods with changing topologies and local updates.
method Unified convergence analysis covering local SGD updates and adaptive network topology.
result Universal convergence rates for smooth problems, interpolating between heterogeneous and iid-data settings.
A distributed SGD method for heterogeneous networks with hubs and workers.
problem Learning in heterogeneous multi-level networks with worker heterogeneity and varying communication.
method Multi-Level Local SGD: distributed SGD with hub-and-spoke paradigm and hub averaging.
result The method converges with error dependent on worker heterogeneity, hub network topology, and iterations.
Unified distributed SGD improves non-convex optimization for large datasets.
problem Bottleneck in scaling SGD for non-convex functions and large datasets.
method Distributed and parallel implementation of SGD (DPSGD) combining asynchronous distribution and lock-free parallelism.
result DPSGD achieves better convergence rate and speed-up with more cores and workers.
Analyzes SGD dynamics in two-layer networks, bridging different regimes.
problem Understanding SGD dynamics in high-dimensional and mean-field settings.
method Rigorous analysis via deterministic low-dimensional description of sufficient statistics.
result Infinite-width dynamics remains close to a low-dimensional subspace.
Proposes Stochastic-Sign SGD for federated learning with theoretical guarantees.
problem Developing efficient, private, and resilient parameter estimation methods for federated learning.
method Introduces Stochastic-Sign SGD, a novel method based on SIGNSGD, which addresses convergence and communication efficiency.
result Demonstrates the effectiveness of Stochastic-Sign SGD through experiments on MNIST and CIFAR-10 datasets.
The gradient noise of SGD is considered to play a central role in the observed strong generalization abilities of deep learning. While past studies confirm that the magnitude and the covariance structure of gradient noise are critical for regularization, it remains unclear whether or not the class of noise distribution…
Paper explores whether gradient normalization can replace clipping for SGD in heavy-tailed noise.
problem Ensuring convergence of SGD in heavy-tailed noise.
method Revisits gradient clipping and normalization, proving their sufficiency and effectiveness.
result Gradient normalization alone is sufficient for nonconvex SGD convergence under smoothness assumptions.
Minibatch SGD outperforms Local SGD in heterogeneous distributed learning.
problem Optimizing a combined convex objective with stochastic gradient estimates from different machines.
method Analysis of Minibatch SGD and Local SGD in a heterogeneous distributed setting.
result Minibatch SGD dominates Local SGD in the heterogeneous distributed setting.
In this note we give a simple proof for the convergence of stochastic gradient (SGD) methods on μ-convex functions under a (milder than standard) L-smoothness assumption. We show that for carefully chosen stepsizes SGD converges after T iterations as $O\left( LR^2 \exp \bigl[-\fracμ{4L}T\bigr] + \frac{σ^2}{μT} \r…
Unified framework for distributed compressed SGD under (L0,L1)-smoothness.
problem Understanding the joint effect of batch noise, adaptivity, and compression in distributed stochastic optimization.
method Developed a unified theoretical framework using SDEs that incorporate curvature-dependent terms.
result Normalizing updates in DCSGD stabilizes convergence, with normalization degree determined by noise structure and landscape regularity.
Analyzes high-dimensional SGD dynamics using DMFT.
problem Understanding the high-dimensional behavior of multi-pass SGD with small batch sizes.
method Derives DMFT equations for high-dimensional SGD dynamics.
result Proves DMFT equations characterize the asymptotic distribution of SGF parameters.
Study reveals convergence properties of SGD with random learning rate.
problem Analyzing convergence of SGD with random learning rate in non-convex optimization.
method Introduced Poisson SGD with random learning rate and used stationary distribution analysis.
result Poisson SGD converges to a stationary distribution and finds global minima in non-convex optimization.
Unified SGD method improves convergence for nested optimization problems.
problem Stochastic nested optimization problems.
method ALTERNATE dESCEN (ALSET) method leveraging hidden smoothness.
result Requires O(ε−2) samples to achieve an ε-stationary point. Local SGD outperforms minibatch SGD for quadratic objectives.
problem Theoretical foundations of local SGD are lacking.
method Proved local SGD strictly dominates minibatch SGD for quadratic objectives and accelerated local SGD is minimax optimal.
result Local SGD does not dominate minibatch SGD in general convex objectives.
Paper shows local SGD outperforms mini-batch SGD under certain conditions.
problem Proving local SGD's superiority in distributed learning with heterogeneous data.
method New lower and upper bounds for local SGD under first-order heterogeneity assumptions.
result Local SGD is min-max optimal under certain conditions, resolving understanding of distributed optimization.
Unified analysis of stochastic gradient methods for convex and smooth optimization.
problem Minimizing composite convex and smooth functions.
method Unified convergence analysis of various stochastic gradient methods.
result Unified convergence rates for a variety of methods including proximal SGD, variance reduced methods, quantization, and coordinate descent.
Overlap-Local-SGD improves distributed SGD by overlapping communication and computation.
problem High communication delay and node slowdown in distributed SGD.
method Adding an anchor model to synchronize local updates and pull them towards the anchor model.
result Overlap-Local-SGD speeds up distributed training and mitigates straggler effects.
New algorithm improves distributed SGD with random sparsification for better convergence and generalization.
problem Communication bottleneck in distributed deep learning.
method Proposes detached error feedback (DEF) algorithm to improve convergence and generalization of communication-efficient distributed SGD.
result Shows better convergence and generalization bounds than existing methods.
Understanding the convergence performance of asynchronous stochastic gradient descent method (Async-SGD) has received increasing attention in recent years due to their foundational role in machine learning. To date, however, most of the existing works are restricted to either bounded gradient delays or convex settings.…
Improved SGD methods converge faster for nonconvex optimization.
problem Nonconvex optimization challenges in machine learning.
method Adaptive SGD with line-search and Polyak stepsizes.
result Unified convergence rates for various nonconvex functions.
Stochastic gradient descent (SGD) is a ubiquitous algorithm for a variety of machine learning problems. Researchers and industry have developed several techniques to optimize SGD's runtime performance, including asynchronous execution and reduced precision. Our main result is a martingale-based analysis that enables us…
Due to its efficiency and ease to implement, stochastic gradient descent (SGD) has been widely used in machine learning. In particular, SGD is one of the most popular optimization methods for distributed learning. Recently, quantized SGD (QSGD), which adopts quantization to reduce the communication cost in SGD-based di…
Integrating adaptive learning rate and momentum techniques into SGD leads to a large class of efficiently accelerated adaptive stochastic algorithms, such as AdaGrad, RMSProp, Adam, AccAdaGrad, \textit{etc}. In spite of their effectiveness in practice, there is still a large gap in their theories of convergences, espec…
New bounds for M-SGD show its error distribution is nearly Gaussian.
problem Understanding the error distribution of M-SGD.
method Proved non-asymptotic bounds for M-SGD in Wasserstein distance.
result Error distribution of M-SGD is approximately Gaussian.
Spectral portfolio theory links neural networks to wealth dynamics via SGD weight matrices.
problem Understanding wealth dynamics from neural network training.
method Direct identification of weight matrices as portfolio allocation matrices, linking SGD forces to portfolio dynamics.
result Spectral properties of SGD weight matrices transition between additive and multiplicative regimes, influencing wealth dynamics.
New method accelerates neural network training by focusing on flat directions.
problem Improving neural network training speed and stability.
method Bulk-SGD, interpolated gradient methods.
result Updates along the Dominant subspace can accelerate convergence but compromise stability.
HybridSGD improves SGD performance by balancing computation and communication.
problem Limited scalability and performance of SGD due to communication costs.
method 2D parallel SGD method (HybridSGD) that trades off between 1D s-step SGD and 1D Federated SGD (FedAvg). result HybridSGD achieves better convergence than FedAvg at similar processor scales and up to 121x speedup over FedAvg.
Communication bottleneck has been identified as a significant issue in distributed optimization of large-scale learning models. Recently, several approaches to mitigate this problem have been proposed, including different forms of gradient compression or computing local models and mixing them iteratively. In this paper…
Unified sign-based compression for federated learning with faster convergence.
problem High communication cost in federated learning with large-scale models.
method Unified noisy perturbation scheme for sign-based compression.
result Achieves faster convergence rate than existing sign-based methods.
CSER improves SGD efficiency by resetting errors and partial synchronization.
problem Limited scalability of Distributed Stochastic Gradient Descent (SGD) due to communication bottlenecks.
method Introduces 'error reset' technique and partial synchronization for gradients and models.
result Proves convergence for smooth non-convex problems and accelerates distributed training significantly.
Adaptive distributed SGD reduces delay in slow workers.
problem Minimizing delay in distributed SGD with stragglers.
method Adaptive policy for varying k to optimize error-runtime trade-off. result Numerical simulations confirm the effectiveness of the adaptive approach.
STL-SGD accelerates Local SGD by gradually increasing communication periods.
problem Reducing communication complexity in distributed SGD.
method Gradually increasing communication periods while decreasing learning rate.
result Significant reduction in communication complexity with improved convergence rates.
Unified convergence analysis of alpha-SVRG under strong convexity.
problem Analyzing the convergence of alpha-SVRG in strongly convex environments.
method Unified convergence rate expression for alpha-SVRG under fixed learning rate, demonstrating faster convergence than SGD and SVRG.
result alpha-SVRG has a faster convergence rate compared to SGD and SVRG under suitable choice of alpha.
Stochastic gradient descent (SGD) is a popular stochastic optimization method in machine learning. Traditional parallel SGD algorithms, e.g., SimuParallel SGD, often require all nodes to have the same performance or to consume equal quantities of data. However, these requirements are difficult to satisfy when the paral…
SGD-trained neural networks generalize well even with adversarial label noise.
problem Generalization of neural networks trained on adversarial label noise.
method Training a one-hidden-layer neural network with SGD on arbitrary width networks.
result SGD-trained networks achieve classification accuracy competitive with the best halfspace over adversarial label noise.
With the increase in the amount of data and the expansion of model scale, distributed parallel training becomes an important and successful technique to address the optimization challenges. Nevertheless, although distributed stochastic gradient descent (SGD) algorithms can achieve a linear iteration speedup, they are l…
Gradient clipping helps private SGD converge despite potential bias.
problem Gradient clipping in private SGD can bias convergence.
method Theoretical analysis and empirical evaluation of gradient clipping effects.
result Gradient clipping can prevent convergence to stationary points and introduces bias.
Non-convex SGD learns halfspaces with adversarial label noise efficiently.
problem Agnostically learning halfspaces in adversarial label noise settings.
method Non-convex SGD optimization for halfspace learning.
result Non-convex SGD achieves misclassification error close to optimal with adversarial noise.
To reduce the long training time of large deep neural network (DNN) models, distributed synchronous stochastic gradient descent (S-SGD) is commonly used on a cluster of workers. However, the speedup brought by multiple workers is limited by the communication overhead. Two approaches, namely pipelining and gradient spar…
Unified analysis of finite weight averaging methods in deep learning.
problem Theoretical explanation of finite weight averaging methods in deep learning.
method Generalized SGD and LAWA, establishing convergence and generalization bounds.
result FWA achieves faster convergence and better generalization compared to SGD.
MindFlayer SGD improves parallel SGD for heterogeneous, random compute times.
problem Minimizing nonconvex functions with heterogeneous, random compute times.
method MindFlayer SGD, designed for stochastic and heterogeneous delays.
result MindFlayer SGD outperforms existing methods in environments with heavy-tailed noise.
Stochastic Gradient Descent (SGD) is the most popular algorithm for training deep neural networks (DNNs). As larger networks and datasets cause longer training times, training on distributed systems is common and distributed SGD variants, mainly asynchronous and synchronous SGD, are widely used. Asynchronous SGD is com…
Machine learning has made tremendous progress in recent years, with models matching or even surpassing humans on a series of specialized tasks. One key element behind the progress of machine learning in recent years has been the ability to train machine learning models in large-scale distributed shared-memory and messa…
We study two types of preconditioners and preconditioned stochastic gradient descent (SGD) methods in a unified framework. We call the first one the Newton type due to its close relationship to the Newton method, and the second one the Fisher type as its preconditioner is closely related to the inverse of Fisher inform…