arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2102.09032cs.DCcs.DS
收藏

快速稳定收敛的一致无锁并行随机梯度下降算法

Consistent Lock-free Parallel Stochastic Gradient Descent for Fast and Stable Convergence

Karl Bäckström, Ivan Walulya, Marina Papatriantafilou, Philippas Tsigas

摘要:

随机梯度下降(SGD)是机器学习(ML)算法中的基本元素。异步并行共享存储器SGD(AsyncSGD),包括无同步算法,例如HOGWILD!由于与同步并行化相比减少了开销,因此在某些上下文中受到了关注。尽管它们会导致陈旧和不一致,但它们在满足光滑,强凸目标和梯度稀疏性的问题上表现出了加速。最近的工作朝着理解并行SGD对于不符合这些强假设的问题的潜力迈出了重要的一步,特别是对于深度学习(DL)。然而,在理解AsyncSGD算法在实践中何时有用,特别是同步和一致性机制如何发挥作用方面,当前文献存在差距。我们重点关注保持一致性的非阻塞同步对SGD收敛的影响,以及对超参数调整的敏感性。我们提出了Leashed-SGD,一个可扩展的算法框架的AsyncSGD的一致性保持实现,采用无锁同步,有效地平衡吞吐量和延迟。我们争论分析的动态算法,内存消耗,线程的进展随着时间的推移,和预期的竞争。我们提供了一个全面的经验评估,验证分析索赔,基准测试拟议的租赁SGD框架,并比较训练多层感知器(MLP)和卷积神经网络(CNN)的基线。我们观察了争用、陈旧性和一致性的关键影响,并展示了与标准的基于锁的AsyncSGD算法和HOGWILD!相比,Leashed-SGD如何在稳定性和时钟收敛时间方面提供显着改进(从20-80%到4倍的改进),同时减少了总的存储器占用空间。

英文摘要:

Stochastic gradient descent (SGD) is an essential element in Machine Learning (ML) algorithms. Asynchronous parallel shared-memory SGD (AsyncSGD), including synchronization-free algorithms, e.g. HOGWILD!, have received interest in certain contexts, due to reduced overhead compared to synchronous parallelization. Despite that they induce staleness and inconsistency, they have shown speedup for problems satisfying smooth, strongly convex targets, and gradient sparsity. Recent works take important steps towards understanding the potential of parallel SGD for problems not conforming to these strong assumptions, in particular for deep learning (DL). There is however a gap in current literature in understanding when AsyncSGD algorithms are useful in practice, and in particular how mechanisms for synchronization and consistency play a role. We focus on the impact of consistency-preserving non-blocking synchronization in SGD convergence, and in sensitivity to hyper-parameter tuning. We propose Leashed-SGD, an extensible algorithmic framework of consistency-preserving implementations of AsyncSGD, employing lock-free synchronization, effectively balancing throughput and latency. We argue analytically about the dynamics of the algorithms, memory consumption, the threads' progress over time, and the expected contention. We provide a comprehensive empirical evaluation, validating the analytical claims, benchmarking the proposed Leashed-SGD framework, and comparing to baselines for training multilayer perceptrons (MLP) and convolutional neural networks (CNN). We observe the crucial impact of contention, staleness and consistency and show how Leashed-SGD provides significant improvements in stability as well as wall-clock time to convergence (from 20-80% up to 4x improvements) compared to the standard lock-based AsyncSGD algorithm and HOGWILD!, while reducing the overall memory footprint.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑