Double Quantization for Communication-Efficient Distributed Optimization
This addresses communication bottlenecks in distributed optimization for machine learning practitioners, though it is incremental as it builds on existing quantization and sparsification techniques.
The paper tackles the problem of high communication overhead in distributed training by proposing double quantization to quantize both model parameters and gradients, resulting in algorithms that effectively save transmitted bits without performance degradation.
Modern distributed training of machine learning models suffers from high communication overhead for synchronizing stochastic gradients and model parameters. In this paper, to reduce the communication complexity, we propose \emph{double quantization}, a general scheme for quantizing both model parameters and gradients. Three communication-efficient algorithms are proposed under this general scheme. Specifically, (i) we propose a low-precision algorithm AsyLPG with asynchronous parallelism, (ii) we explore integrating gradient sparsification with double quantization and develop Sparse-AsyLPG, (iii) we show that double quantization can also be accelerated by momentum technique and design accelerated AsyLPG. We establish rigorous performance guarantees for the algorithms, and conduct experiments on a multi-server test-bed to demonstrate that our algorithms can effectively save transmitted bits without performance degradation.