Demystifying Why Local Aggregation Helps: Convergence Analysis of Hierarchical SGD

• URL: http://arxiv.org/abs/2010.12998v4
• Date: Thu, 11 Apr 2024 17:05:18 GMT
• Title: Demystifying Why Local Aggregation Helps: Convergence Analysis of Hierarchical SGD
• Abstract summary: Hierarchical SGD (H-SGD) has emerged as a new distributed SGD algorithm for multi-level communication networks. In H-SGD, before each global aggregation, workers send their updated local models to local servers for aggregations. Despite recent research efforts, global convergence still lacks theoretical understanding.
• Score: 39.75009345804017
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Hierarchical SGD (H-SGD) has emerged as a new distributed SGD algorithm for multi-level communication networks. In H-SGD, before each global aggregation, workers send their updated local models to local servers for aggregations. Despite recent research efforts, the effect of local aggregation on global convergence still lacks theoretical understanding. In this work, we first introduce a new notion of "upward" and "downward" divergences. We then use it to conduct a novel analysis to obtain a worst-case convergence upper bound for two-level H-SGD with non-IID data, non-convex objective function, and stochastic gradient. By extending this result to the case with random grouping, we observe that this convergence upper bound of H-SGD is between the upper bounds of two single-level local SGD settings, with the number of local iterations equal to the local and global update periods in H-SGD, respectively. We refer to this as the "sandwich behavior". Furthermore, we extend our analytical approach based on "upward" and "downward" divergences to study the convergence for the general case of H-SGD with more than two levels, where the "sandwich behavior" still holds. Our theoretical results provide key insights of why local aggregation can be beneficial in improving the convergence of H-SGD.

Related papers

• What's in a Smoothness Constant? Tighter Rates for Local SGD with Bounded Second-order Heterogeneity [39.01981872811199]
  Local SGD, also known as Federated Averaging, is a widely used distributed optimization algorithm.<n>Recent work by [Patel et al., 2025] shows that a bounded second-order heterogeneity assumption captures the efficiency of Local SGD for strongly convex objectives.<n>We prove this conjecture by establishing an improved convergence guarantee for Local SGD on general convex objectives under bounded second-order heterogeneity.
  arXiv  Detail & Related papers  (2026-07-16T08:56:45Z)
• On the Convergence of (Stochastic) Gradient Descent for Kolmogorov--Arnold Networks [56.78271181959529]
  Kolmogorov--Arnold Networks (KANs) have gained significant attention in the deep learning community. Empirical investigations demonstrate that KANs optimized via gradient descent (SGD) are capable of achieving near-zero training loss.
  arXiv  Detail & Related papers  (2024-10-10T15:34:10Z)
• The Limits and Potentials of Local SGD for Distributed Heterogeneous Learning with Intermittent Communication [37.210933391984014]
  Local SGD is a popular optimization method in distributed learning, often outperforming other algorithms in practice. We provide new lower bounds for local SGD under existing first-order data heterogeneity assumptions. We also show the min-max optimality of accelerated mini-batch SGD for several problem classes.
  arXiv  Detail & Related papers  (2024-05-19T20:20:03Z)
• From Gradient Flow on Population Loss to Learning with Stochastic Gradient Descent [50.4531316289086]
  Gradient Descent (SGD) has been the method of choice for learning large-scale non-root models. An overarching paper is providing general conditions SGD converges, assuming that GF on the population loss converges. We provide a unified analysis for GD/SGD not only for classical settings like convex losses, but also for more complex problems including Retrieval Matrix sq-root.
  arXiv  Detail & Related papers  (2022-10-13T03:55:04Z)
• Linearly Converging Error Compensated SGD [11.436753102510647]
  We propose a unified analysis of variants of distributed SGD with arbitrary compressions and delayed updates. Our framework is general enough to cover different variants of quantized SGD, ErrorCompensated SGD and SGD with delayed updates. We develop new variants of SGD that combine variance reduction or arbitrary sampling with error feedback and quantization.
  arXiv  Detail & Related papers  (2020-10-23T10:46:31Z)
• O(1) Communication for Distributed SGD through Two-Level Gradient Averaging [0.0]
  We introduce a strategy called two-level gradient averaging (A2SGD) to consolidate all gradients down to merely two local averages per worker. Our theoretical analysis shows that A2SGD converges similarly like the default distributed SGD algorithm.
  arXiv  Detail & Related papers  (2020-06-12T18:20:52Z)
• Minibatch vs Local SGD for Heterogeneous Distributed Learning [28.80878557506603]
  We argue that Minibatch SGD dominates all existing analysis of Local SGD in this setting. We present the first upper bound for Local SGD that improves over Minibatch SGD in a non-homogeneous regime.
  arXiv  Detail & Related papers  (2020-06-08T16:40:49Z)
• Global Distance-distributions Separation for Unsupervised Person Re-identification [93.39253443415392]
  Existing unsupervised ReID approaches often fail in correctly identifying the positive samples and negative samples through the distance-based matching/ranking. We introduce a global distance-distributions separation constraint over the two distributions to encourage the clear separation of positive and negative samples from a global view. We show that our method leads to significant improvement over the baselines and achieves the state-of-the-art performance.
  arXiv  Detail & Related papers  (2020-06-01T07:05:39Z)
• Breaking (Global) Barriers in Parallel Stochastic Optimization with Wait-Avoiding Group Averaging [48.99717153937717]
  We present WAGMA-SGD, a wait-avoiding subgroup that reduces global communication via weight exchange.<n>We train ResNet-50 on ImageNet; Transformer for machine translation; and deep reinforcement learning for navigation at scale.<n>Compared with state-of-the-art decentralized SGD variants, WAGMA-SGD significantly improves training throughput.
  arXiv  Detail & Related papers  (2020-04-30T22:11:53Z)
• Detached Error Feedback for Distributed SGD with Random Sparsification [98.98236187442258]
  Communication bottleneck has been a critical problem in large-scale deep learning. We propose a new distributed error feedback (DEF) algorithm, which shows better convergence than error feedback for non-efficient distributed problems. We also propose DEFA to accelerate the generalization of DEF, which shows better bounds than DEF.
  arXiv  Detail & Related papers  (2020-04-11T03:50:59Z)
• A Unified Theory of Decentralized SGD with Changing Topology and Local Updates [70.9701218475002]
  We introduce a unified convergence analysis of decentralized communication methods. We derive universal convergence rates for several applications. Our proofs rely on weak assumptions.
  arXiv  Detail & Related papers  (2020-03-23T17:49:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.