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.
Improved Local SGD convergence for general convex objectives with bounded second-order heterogeneity.
problem Understanding when and why Local SGD outperforms alternatives in distributed optimization.
method Established improved convergence guarantees for Local SGD on general convex objectives under bounded second-order heterogeneity.
result Upper bounds for Local SGD are nearly tight, providing a sharper convergence theory.
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.
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.
Improved local SGD analysis for identical and heterogeneous data.
problem Analyzing local SGD performance in identical and heterogeneous data regimes.
method Developed a new variance measure specific to local SGD and improved existing theory.
result Tightened stepsize and number of local iterations for optimal performance.
Local SGD outperforms conventional methods in LLM training.
problem Training large language models on distributed devices.
method Local SGD algorithm applied to distributed training.
result Local SGD achieves competitive results in LLM training.
This paper analyzes Local SGD for federated learning, achieving both statistical and communication efficiency.
problem Statistical estimation and inference in federated learning with decentralized data.
method Local SGD, a multi-round estimation procedure using intermittent communication.
result Local SGD achieves both statistical efficiency and communication efficiency.
Qsparse-local-SGD reduces communication in large-scale learning models.
problem Communication bottleneck in distributed optimization of large-scale models.
method Combines sparsification, quantization, and local computation with error compensation.
result Converges at the same rate as vanilla distributed SGD for many sparsifiers and quantizers.
The paper provides Gaussian approximations for decentralized Federated Learning.
problem Lack of asymptotic statistical guarantees for local SGD in Federated Learning.
method Two generalized Gaussian approximation results for local SGD trajectories.
result Valid multiplier bootstrap procedures and Gaussian bootstrap-based tests for detecting adversarial attacks.
New analysis shows Local SGD can achieve error scaling with only fixed number of communications.
problem Speeding up SGD by parallelizing across multiple workers with reduced communication overhead.
method Proposed and analyzed Local SGD method with a fixed number of communications independent of the number of steps.
result Achieves an error scaling as 1/(nT) with only a fixed number of communications (Ω(n)).
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.
VRL-SGD reduces communication complexity in non-identical data settings.
problem Training machine learning models with non-identical data distribution.
method VRL-SGD, which eliminates gradient variance dependency and achieves linear speedup with lower communication complexity.
result VRL-SGD reduces communication complexity from $O(T^{ rac{3}{4}} N^{ rac{3}{4}})$ to $O(T^{ rac{1}{2}} N^{ rac{3}{2}})$.
New insights into optimizing Local SGD's outer optimizer for faster convergence.
problem Understanding the impact of outer optimizer and its hyperparameters in Local SGD.
method Analyzing convergence guarantees with new outer learning rates and momentum.
result Tuning the outer learning rate can improve convergence and handle inner learning rate ill-tuning.
Local SGD with periodic averaging achieves faster convergence with less communication.
problem Communication overhead in distributed optimization.
method Local SGD with periodic averaging, Polyak-Łojasiewicz condition, adaptive synchronization.
result Local SGD can achieve linear speed up with fewer communication rounds, especially for non-strongly convex functions.
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.
Mini-batch stochastic gradient methods (SGD) are state of the art for distributed training of deep neural networks. Drastic increases in the mini-batch sizes have lead to key efficiency and scalability gains in recent years. However, progress faces a major roadblock, as models trained with large batches often do not ge…
Local SGD proves efficient in overparameterized linear regression.
problem Efficiently learning overparameterized linear models in distributed settings.
method Distributed SGD (DSGD) with overparameterized models.
result Excess risk of SGD is smaller than ridge regression in the same sample complexity.
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.
New algorithm reduces communication in distributed SGD, improving efficiency.
problem Slow communication rounds bottleneck synchronous mini-batch SGD convergence.
method Proposes non-asymptotic error analysis for Local-SGD, comparing to averaging methods.
result Local-SGD reduces communication by a factor of O(√T/P^(3/2)) for large step sizes.
Sharp bounds established for Federated Averaging (FedAvg), improving convergence rates.
problem Undetermined convergence rate of Federated Averaging (FedAvg) in Federated Learning.
method Developed novel iterate bias concept and proved sharp bounds on it, leading to improved convergence results.
result Lower bounds for FedAvg match existing upper bounds, showing no improvable capacity.
New convergence bounds for shuffling-based SGD methods in distributed learning.
problem Analyzing the performance of shuffling-based variants of SGD in distributed learning.
method Study of minibatch and local Random Reshuffling methods, proving convergence bounds and lower bounds.
result Shuffling-based variants converge faster than with-replacement sampling methods, and the bounds are tight.
New methods for federated learning reduce communication costs.
problem Efficiently solving optimization problems in a distributed setting.
method Developed two strategies for achieving consensus in federated learning: fixed number of local steps and randomized computations.
result Convergence analysis and experiments show benefits of the proposed methods.
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.
The thesis clarifies when local updates outperform centralized methods in heterogeneous data environments.
problem Understanding when local updates are more effective than centralized or mini-batch methods in distributed optimization.
method Fine-grained consensus-error-based analysis framework, focusing on bounded second-order heterogeneity and third-order smoothness.
result Local updates outperform centralized or mini-batch methods under realistic models of data heterogeneity.
A new method reduces communication costs in distributed learning.
problem Reduces communication bottlenecks in distributed learning.
method Local SGD with communication-computation overlap and delay-corrected sparse model averaging.
result Theoretical convergence guarantees for smooth non-convex objectives.
FedAc accelerates Federated Averaging for distributed optimization.
problem Efficiently optimizing distributed machine learning models.
method Federated Accelerated Stochastic Gradient Descent (FedAc) using a potential-based perturbed iterate analysis.
result FedAc achieves faster convergence and lower communication costs than previous methods.
SQuARM-SGD improves decentralized SGD efficiency with momentum.
problem Efficient decentralized training of large-scale models over networks.
method Fixed local SGD steps with Nesterov's momentum, sparsified and quantized updates, locally computed triggering criterion.
result Convergence rate matches vanilla SGD, momentum improves test performance.
Adaptive batch sizes improve local gradient methods in distributed training.
problem Communication bottlenecks in distributed deep learning.
method Adaptive batch size strategies for local gradient methods.
result Adaptive batch sizes reduce minibatch gradient variance and improve training efficiency.
Partial model averaging improves Federated Learning performance.
problem Periodic model averaging causes significant model discrepancy in Federated Learning.
method Proposes a partial model averaging framework that encourages local models to stay close to each other.
result Partial averaging achieves up to 2.2% higher validation accuracy than full averaging.
New optimization for federated learning with local models.
problem Training models with private data from multiple devices.
method Proposes a new optimization formulation and efficient SGD variants.
result Local steps can improve communication for heterogeneous data.
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.
SlowMo improves distributed SGD by periodically synchronizing and using momentum, leading to better accuracy.
problem Reducing communication overhead in distributed training while maintaining model accuracy.
method Periodically synchronizing workers and using momentum after multiple iterations of a base optimization algorithm.
result SlowMo consistently yields improvements in optimization and generalization performance relative to the base optimizer.
Improved SGD rates with delayed and compressed gradients.
problem Convergence rates of SGD with delayed and compressed gradients.
method Analyzed SGD with delayed updates on smooth quasi-convex and non-convex functions, derived non-asymptotic rates.
result Convergence rates are affected by noise but not by delay, leading to optimal rates.
Hogwild! adapts to distributed data by varying mini-batch sizes.
problem Efficiently parallelize SGD over distributed local data sets.
method Asynchronous SGD with varying mini-batch sizes, aggregated by an aggregator.
result Improved convergence for heterogeneous data, reducing communication rounds.
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.
New method improves convergence in federated learning for nonconvex problems.
problem Optimizing global objective in distributed learning with non-i.i.d. data.
method Generalized local stochastic and full gradient descent with periodic averaging.
result Demonstrates convergence rates for nonconvex federated optimization.
Local Gradient Descent with local steps converges to the centralized model in the interpolation regime.
problem Understanding the implicit bias of Local Gradient Descent in the interpolation regime.
method Analyzing the implicit bias of Local Gradient Descent for classification tasks with linearly separable data.
result The aggregated global model from Local-GD converges exactly to the centralized model in the interpolation regime.
New algorithm reduces FL sample and communication costs.
problem Optimizing FL for minimal samples and rounds.
method Stochastic two-sided momentum algorithm.
result Achieves near-optimal sample and communication complexities.
Improved federated learning over-the-air reduces noise impact.
problem Noise in OTA FL degrades model accuracy.
method Developed COTAF algorithm with precoding and scaling.
result COTAF achieves similar convergence rate to error-free channels.