Helix: Holistic Optimization for Accelerating Iterative Machine Learning
Helix: Holistic Optimization for Accelerating Iterative Machine Learning
复制标题
DOI：
10.14778/3297753.3297763
复制
发表时间：
2018-12
期刊：
影响因子：
--
通讯作者：
Doris Xin;Stephen Macke;Litian Ma;Jialin Liu;Shuchen Song;Aditya G. Parameswaran
中科院分区：
文献类型：
--
作者：
Doris Xin;Stephen Macke;Litian Ma;Jialin Liu;Shuchen Song;Aditya G. Parameswaran
机器学习工作流开发是一个反复试验的过程：开发人员通过测试小的修改来迭代工作流，直到达到期望的精度。不幸的是，现有的机器学习系统狭隘地专注于模型训练——这只是整个开发时间的一小部分——而忽略了解决迭代开发问题。我们提出了H elix，一种机器学习系统，它优化了跨迭代的执行——智能地缓存和复用，或者适当地重新计算中间结果。H elix在其Scala领域特定语言（DSL）中涵盖了各种各样的应用需求，用简洁的语法定义了数据预处理、模型规范和学习的统一流程。我们证明了复用问题可以转化为一个最大流问题，而缓存问题是NP难问题。我们为后者开发了有效的轻量级启发式算法。实证评估表明，H elix不仅能够在一个统一的工作流中处理各种各样的用例，而且速度更快，在自然语言处理、计算机视觉、社会科学和自然科学的四个实际应用中，与最先进的系统（如DeepDive或KeystoneML）相比，运行时间最多可减少19倍。
Machine learning workflow development is a process of trial-and-error: developers iterate on workflows by testing out small modifications until the desired accuracy is achieved. Unfortunately, existing machine learning systems focus narrowly on model training---a small fraction of the overall development time---and neglect to address iterative development. We propose H elix , a machine learning system that optimizes the execution across iterations ---intelligently caching and reusing, or recomputing intermediates as appropriate. H elix captures a wide variety of application needs within its Scala DSL, with succinct syntax defining unified processes for data preprocessing, model specification, and learning. We demonstrate that the reuse problem can be cast as a M ax -F low problem, while the caching problem is NP-H ard . We develop effective lightweight heuristics for the latter. Empirical evaluation shows that H elix is not only able to handle a wide variety of use cases in one unified workflow but also much faster, providing run time reductions of up to 19x over state-of-the-art systems, such as DeepDive or KeystoneML, on four real-world applications in natural language processing, computer vision, social and natural sciences.