The proliferation of massive datasets has led to significant interests in distributed algorithms for solving large-scale machine learning problems. However, the communication overhead is a major bottleneck that hampers the scalability of distributed machine learning systems. In this paper, we design two communication-efficient algorithms for distributed learning tasks. The first one is named EF-SIGNGD, in which we use the 1-bit (sign-based) gradient quantization method to save the communication bits. Moreover, the error feedback technique, i.e., incorporating the error made by the compression operator into the next step, is employed for the convergence guarantee. The second algorithm is called LE-SIGNGD, in which we introduce a well-designed lazy gradient aggregation rule to EF-SIGNGD that can detect the gradients with small changes and reuse the outdated information. LE-SIGNGD saves communication costs both in transmitted bits and communication rounds. Furthermore, we show that LE-SIGNGD is convergent under some mild assumptions. The effectiveness of the two proposed algorithms is demonstrated through experiments on both real and synthetic data.
21 May 2025 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Stochastic gradient descent (SGD) performed in an asynchronous manner plays a crucial role in training large-scale machine learning models. However, the generalization performance of asynchronous delayed SGD, which is an essential metric for assessing machine learning algorithms, has rarely been explored. Existing generalization error bounds are rather pessimistic and cannot reveal the correlation between asynchronous delays and generalization. In this paper, we investigate sharper generalization error bound for SGD with asynchronous delay $\tau$. Leveraging the generating function analysis tool, we first establish the average stability of the delayed gradient algorithm. Based on this algorithmic stability, we provide upper bounds on the generalization error of $\widetilde{\mathcal {O}}(\frac{T-\tau }{n\tau })$ and $\widetilde{\mathcal {O}}(\frac{1}{n})$ for quadratic convex and strongly convex problems, respectively, where $T$ refers to the iteration number and $n$ is the amount of training data. Our theoretical results indicate that asynchronous delays reduce the generalization error of the delayed SGD algorithm. Analogous analysis can be generalized to the random delay setting, and the experimental results validate our theoretical findings.
Subspace clustering has been widely applied to detect meaningful clusters in high-dimensional data spaces. And the sparse subspace clustering (SSC) obtains superior clustering performance by solving a relaxed ℓ0-minimization problem with ℓ1-norm. Although the use of ℓ1-norm instead of the ℓ0 one can make the object function convex, it causes large errors on large coefficients in some cases. In this paper, we study the sparse subspace clustering algorithm based on a nonconvex modeling formulation. Specifically, we introduce a nonconvex pseudo-norm that makes a better approximation to the ℓ0-minimization than the traditional ℓ1-minimization framework and consequently finds a better affinity matrix. However, this formulation makes the optimization task challenging due to that the traditional alternating direction method of multipliers (ADMM) encounters troubles in solving the nonconvex subproblems. In view of this, the reweighted techniques are employed in making these subproblems convex and easily solvable. We provide several guarantees to derive the convergence results, which proves that the nonconvex algorithm is globally convergent to a critical point. Experiments on two real-world problems of motion segmentation and face clustering show that our method outperforms state-of-the-art techniques.
Gradient-based optimization methods implemented on distributed computing architectures are increasingly used to tackle large-scale machine learning applications. A key bottleneck in such distributed systems is the high communication overhead for exchanging information, such as stochastic gradients, between workers. The inherent causes of this bottleneck are the frequent communication rounds and the full model gradient transmission in every round. In this study, we present SASG, a communication-efficient distributed algorithm that enjoys the advantages of sparse communication and adaptive aggregated stochastic gradients. By dynamically determining the workers who need to communicate through an adaptive aggregation rule and sparsifying the transmitted information, the SASG algorithm reduces both the overhead of communication rounds and the number of communication bits in the distributed system. For the theoretical analysis, we introduce an important auxiliary variable and define a new Lyapunov function to prove that the communication-efficient algorithm is convergent. The convergence result is identical to the sublinear rate of stochastic gradient descent, and our result also reveals that SASG scales well with the number of distributed workers. Finally, experiments on training deep neural networks demonstrate that the proposed algorithm can significantly reduce communication overhead compared to previous methods.
In the domain of computer architecture, Federated Learning (FL) is a paradigm of distributed machine learning in edge systems. However, the systems’ Non-Independent and Identically Distributed (Non-IID) data negatively affect the convergence efficiency of the global model, since only a subset of these data samples is beneficial for accelerating model convergence. In pursuit of this subset, a reliable approach involves determining a measure of validity to rank the samples within the dataset. In this paper, we propose the BHerd strategy, which selects a beneficial herd of local gradients to accelerate the convergence of the FL model. Specifically, we map the distribution of the local dataset to the local gradients and use the Herding strategy to obtain a permutation of the set of gradients, where the more advanced gradients in the permutation are closer to the average of the set of gradients. These top portions of the gradients will be selected and sent to the server for global aggregation. We conduct experiments on different datasets, models, and scenarios by building a prototype system, and experimental results demonstrate that our BHerd strategy is effective in selecting beneficial local gradients to mitigate the effects brought by the Non-IID dataset.