LAG: Lazily Aggregated Gradient for Communication-Efficient Distributed Learning
This work addresses communication efficiency for distributed learning systems, particularly in heterogeneous data settings, representing an incremental improvement over existing gradient methods.
The paper tackles the problem of high communication costs in distributed machine learning by introducing LAG, a gradient method that adaptively reuses outdated gradients to reduce communication and computation. The result is a method that maintains the same convergence rate as batch gradient descent across various convexity conditions and reduces communication rounds, with numerical experiments showing significant communication reduction compared to alternatives.
This paper presents a new class of gradient methods for distributed machine learning that adaptively skip the gradient calculations to learn with reduced communication and computation. Simple rules are designed to detect slowly-varying gradients and, therefore, trigger the reuse of outdated gradients. The resultant gradient-based algorithms are termed Lazily Aggregated Gradient --- justifying our acronym LAG used henceforth. Theoretically, the merits of this contribution are: i) the convergence rate is the same as batch gradient descent in strongly-convex, convex, and nonconvex smooth cases; and, ii) if the distributed datasets are heterogeneous (quantified by certain measurable constants), the communication rounds needed to achieve a targeted accuracy are reduced thanks to the adaptive reuse of lagged gradients. Numerical experiments on both synthetic and real data corroborate a significant communication reduction compared to alternatives.