Abstraction library for enabling scalable distributed machine learning

The invention discloses an abstract library for enabling scalable distributed machine learning. One embodiment provides a non-transitory machine-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations that include providing an interface for defining a neural network using machine learning domain-specific terms, the interface enables selection of a neural network topology and abstracting low-layer communication details for distributed training of the neural network. 本申请公开了用于使得能够进行可扩展分布式机器学习的抽象库。一个实施例提供了一种存储有指令的非暂态机器可读介质，所述指令当由一个或多个处理器执行时使所述一个或多个处理器执行包括以下各项的操作：提供用于使用机器学习领域特定术语来定义神经网络的界面，其中，所述界面使得能够选择神经网络拓扑并且抽象出所述神经网络的分布式训练的低层通信细节。